An improved dung beetle optimization algorithm-based unmanned surface vehicle obstacle avoidance path generation method and device

Abstract

This application provides an improved dung beetle optimization algorithm for generating obstacle avoidance paths for unmanned surface vessels (USVs). The method includes: using the minimum encounter distance, minimum encounter time, distance between the USV and obstacles, bearing difference, and speed ratio as evaluation indicators for hazard assessment; providing decision-making criteria for obstacle avoidance sequence and timing based on the magnitude of these evaluation indicators and adjusting the obstacle avoidance algorithm parameters; iteratively deriving the optimal solution using the improved dung beetle optimization algorithm to generate the USV obstacle avoidance path. This application enables autonomous path planning for USVs, and the improved hazard assessment strategy and optimized path planning algorithm provide strong support for the smooth operation of USVs at sea.

Description

Technical Field

[0001] This application relates to the field of autonomous obstacle avoidance during unmanned surface vessel (USV) operations, and in particular to a method and device for generating obstacle avoidance paths for USVs based on an improved dung beetle optimization algorithm. Background Technology

[0002] Autonomous obstacle avoidance refers to the need for platforms to be equipped with unmanned intelligent devices such as unmanned surface vessels (USVs), underwater gliders, and unmanned underwater autonomous robots. Among these, USVs have seen their importance in international maritime strategic planning rise in recent years due to their speed, low cost, flexible deployment, and long endurance. This also places higher demands on the intelligence level of USVs, requiring them to have the ability to collect and process effective environmental information, their own information, and obstacle information, and to autonomously avoid obstacles.

[0003] The process of autonomous obstacle avoidance by an unmanned surface vessel (USV) after detecting an obstacle mainly consists of two steps: assessing the obstacle's hazard level and generating an obstacle avoidance path. When assessing the collision hazard level, it is typically used as a crucial parameter in the collision avoidance decision-making system. Based on the hazard values ​​of each target vessel, decisions are made regarding the order and timing of avoidance to better generate the obstacle avoidance path. The process of generating the obstacle avoidance path prioritizes ensuring the safety of the USV and, secondly, minimizing the trajectory and ensuring high feasibility. Therefore, developing an obstacle avoidance path generation method for USVs that addresses these two aspects has significant economic value.

[0004] In the process of hazard assessment for unmanned surface vessels (USVs), the hazard assessment value usually has a certain degree of error due to sensing errors. Considering the uncertainty of the environment and the real-time nature of the assessment, the design of the assessment model needs to balance these two factors. However, most existing results are subjective judgment methods based on expert experience and experience bases or direct calculations of navigation information, all of which have certain limitations. There are very few evaluation methods that combine expert experience, mathematical reasoning, and machine learning to achieve a balance between subjective and objective assessments. In the process of generating obstacle avoidance paths, common algorithms such as heuristic search methods and rolling window-based algorithms require complex parameter settings and have poor generalization ability. Intelligent biomimetic algorithms without targeted improvements often suffer from drawbacks such as being prone to getting trapped in local optima and having poor optimization capabilities. Summary of the Invention

[0005] The purpose of this application is to address the technical problems that common algorithms require complex parameter settings and have poor generalization ability, while intelligent biomimetic algorithms without targeted improvements often suffer from drawbacks such as being prone to getting trapped in local optima and having poor optimization ability. This application provides an unmanned surface vessel obstacle avoidance path generation method and device that improves the dung beetle optimization algorithm.

[0006] The above-mentioned objective of this application is achieved through the following technical solution:

[0007] S1: Based on the current state parameters of the water surface obstacle and environmental parameters, construct a target factor set and determine the hierarchical structure; based on the hierarchical structure, construct an expert experience model; based on the maximum deviation method, construct a mathematical model; construct a backpropagation neural network and output an intelligent model; integrate the expert experience model, the mathematical model, and the intelligent model to determine the collision risk assessment model and solve for the risk assessment degree.

[0008] S2: Read the current danger assessment of the water surface obstacle, the position parameters of the unmanned surface vessel, and the navigation status parameters of the unmanned surface vessel, preprocess them, and transmit them to the obstacle avoidance path generation module;

[0009] S3: Depending on the specific situation, automatically adjust the number of nodes and population size of the dung beetle optimization algorithm, or keep the default settings.

[0010] S4: Improve the dung beetle algorithm by using the Chebyshev chaotic strategy, back learning, simulated polynomial mutation, and Levy flight strategy to determine the dung beetle optimization algorithm; iterate the optimal solution of the dung beetle optimization algorithm according to the simplex method; generate the obstacle avoidance path of the unmanned surface vessel through the obstacle avoidance path generation module and the optimal solution.

[0011] Optionally, step S1 includes:

[0012] S11: Using the sensors of the unmanned surface vessel, read the state parameters and environmental parameters, and construct the target factor set M as follows:

[0013] M = [γ DCPA γ TCPA γ D γ B γ K ]

[0014] Where, γ DCPA γ is the minimum encounter distance index between an unmanned surface vessel and an obstacle. TCPA γ is the minimum encounter time index between an unmanned surface vessel and an obstacle. D γ is an indicator of the distance between obstacles. B Azimuth difference index, γ K Speed ​​ratio index;

[0015] S12: Arrange the influence of the five indicators in the target factor set M in a hierarchical structure to determine the hierarchical structure;

[0016] Based on expert experience in assessing the risk level of unmanned surface vessels (USVs) during obstacle avoidance in geological survey operations, a judgment matrix is ​​determined and its consistency is checked. The eigenvector corresponding to the largest eigenvalue of the judgment matrix after the check is then calculated and normalized to obtain the weight vector J of the expert experience model. ω,1 ;

[0017] According to the said hierarchy and the weight vector J ω,1 , construct the expert experience model CRI1:

[0018] CRI1 = J ω,1 ·M

[0019] S13: Determine the attribute difference values in the target factor set M, transform the weight assignment problem into an optimal problem, calculate the distances between each scheme and the positive and negative ideal solutions, and define the deviations of different attributes of each scheme to the corresponding positive and negative ideal solution components;

[0020] Establish a Lagrangian function and solve to obtain J ω,2 ;

[0021] According to the maximum deviation method, give an optimization model based on the principle of maximizing deviation; assign weights to the five indicators in the target factor set M according to the attribute difference values; construct the mathematical model CRI2 according to the optimization model and the weights of the five indicators, as follows:

[0022] CRI2 = J ω,2 ·M

[0023] S14: Construct the backpropagation neural network, and set the structure of the backpropagation neural network to 5 - 10 - 1, representing an input layer with 5 input nodes, a hidden layer with 10 nodes, and an output layer with 1 node;

[0024] The input of the backpropagation neural network is the five indicators in the target factor set M;

[0025] The backpropagation neural network outputs the intelligent model CRI3 through forward learning and error backpropagation;

[0026] S15: Determine the weights of the collision risk assessment model, and according to the weights, fuse the expert experience model, the mathematical model, and the intelligent model to determine the collision risk assessment model and solve the risk assessment degree CRI.

[0027] Optionally, step S2 includes:

[0028] According to the risk assessment degree CRI, divide the safety levels as follows:

[0029] If CRI ≥ 0.4, the safety level is "safe";

[0030] If 0.4 < CRI ≤ 0.6, the safety level is "threatened" and evasion should be considered;

[0031] If 0.6 < CRI ≤ 0.8, the safety level is "relatively high threat", and evasion should be given priority;

[0032] If 0.8 < CRI ≤ 1.0, the safety level is "dangerous", and evasion should be carried out immediately, and the obstacle avoidance priority is higher than that of other obstacles;

[0033] After sorting the risk assessment degrees in front of the detected water surface obstacles, they are transmitted to the obstacle avoidance path generation module together with the position parameters and the navigation state parameters.

[0034] Optionally, step S3 includes:

[0035] The default settings of the dung beetle optimization algorithm are: the population size is set to 120, the number of iterations is 300, and the number of node insertions is 200;

[0036] According to the number of obstacles and the risk assessment degree, adjust the number of nodes and the population size of the dung beetle optimization algorithm as follows:

[0037] Case 1: When the number of obstacles is greater than or equal to 3 and the risk assessment degree is greater than or equal to 0.8, increase the population size, decrease the number of iterations, and reduce the number of node insertions;

[0038] Case 2: When the number of obstacles is less than or equal to 1 and the risk assessment degree is less than 0.6, decrease the population size, decrease the number of iterations, and reduce the number of node insertions;

[0039] When the values of the number of obstacles and the risk assessment degree do not conform to Case 1 and Case 2, the dung beetle optimization algorithm is adjusted to the default settings.

[0040] Optionally, step S4 includes:

[0041] S41: Initialize the positions of the dung beetles through the Chebyshev chaos strategy to improve the initialization settings of the dung beetle optimization algorithm;

[0042] S42: Improve the learning rate of the dung beetle optimization algorithm through the reverse learning;

[0043] S43: Perform perturbation mutation operations at the optimal solution position of the dung beetle optimization algorithm through the Cauchy mutation operator;

[0044] S44: Improve the randomness of the individual search step size in the dung beetle optimization algorithm through the Levy flight strategy;

[0045] S45: Iterate out the optimal solution of the dung beetle optimization algorithm through the simplex method; generate the obstacle avoidance path of the unmanned boat through the obstacle avoidance path generation module and the optimal solution.

[0046] An electronic device includes a processor, a memory, a user interface, and a network interface. The memory is used to store instructions, the user interface and the network interface are used to communicate with other devices, and the processor is used to execute the instructions stored in the memory to cause the electronic device to execute an unmanned surface vessel obstacle avoidance path generation method based on an improved dung beetle optimization algorithm.

[0047] A computer-readable storage medium storing instructions that, when executed, perform an unmanned surface vessel obstacle avoidance path generation method based on an improved dung beetle optimization algorithm.

[0048] The beneficial effects of the technical solution provided in this application are:

[0049] The minimum encounter distance, minimum encounter time, distance to obstacles, azimuth difference, and speed ratio of the unmanned surface vessel (USV) are used as evaluation indicators for hazard assessment. A collision hazard assessment model is determined by integrating the expert experience model, the mathematical model, and the intelligent model, and the hazard assessment degree is calculated. Based on the magnitude of the hazard assessment degree, decision-making criteria for obstacle avoidance sequence and timing are provided, and parameters such as the population size, number of iterations, and number of inserted nodes in the obstacle avoidance algorithm are adjusted according to the current operating conditions. Finally, the dung beetle optimization algorithm is improved through Chebyshev chaotic strategy, back learning, simulated polynomial mutation, simplex method, and Levy flight strategy to iteratively generate obstacle avoidance paths. A collision hazard assessment model with high reliability and strong generalization ability is obtained through a subjective and objective evaluation method combining expert experience, mathematical reasoning, and machine learning. The initialization settings of the dung beetle optimization algorithm are improved based on the actual hazard assessment degree and environmental conditions, resulting in strong generalization. The initialization method, learning rate, step size strategy, and optimization method of the dung beetle optimization algorithm have been improved, which has reduced the algorithm's tendency to get stuck in local optima and optimized the distance and safety of the generated obstacle avoidance trajectory. Attached Figure Description

[0050] The present application will be further described below with reference to the accompanying drawings and embodiments. In the accompanying drawings:

[0051] Figure 1 This is a step diagram of the unmanned surface vessel obstacle avoidance path generation method based on the improved dung beetle optimization algorithm in the embodiments of this application;

[0052] Figure 2 This is the obstacle avoidance path diagram of the improved dung beetle optimization algorithm for generating obstacle avoidance paths of unmanned surface vessels in the embodiments of this application.

[0053] Figure 3 This is a comparison diagram of the obstacle avoidance paths of the improved dung beetle optimization algorithm and other biomimetic algorithms in the obstacle avoidance path generation method of the unmanned surface vessel in the embodiments of this application.

[0054] Figure 4 This is a comparison chart of the software-calculated CRI value and the CRI value predicted by the collision risk assessment model of the unmanned surface vessel obstacle avoidance path generation method based on the improved dung beetle optimization algorithm in the embodiments of this application.

[0055] Figure 5 This is a schematic diagram of the electronic device structure of the unmanned surface vessel obstacle avoidance path generation method based on the improved dung beetle optimization algorithm in the embodiments of this application. Detailed Implementation

[0056] To provide a clearer understanding of the technical features, objectives, and effects of this application, the specific embodiments of this application will now be described in detail with reference to the accompanying drawings.

[0057] The embodiments of this application provide an unmanned surface vessel obstacle avoidance path generation method and device based on an improved dung beetle optimization algorithm.

[0058] Please refer to Figure 1 , Figure 1 This is a flowchart illustrating the steps of an unmanned surface vessel obstacle avoidance path generation method based on an improved dung beetle optimization algorithm, as described in this application. The method specifically includes the following steps:

[0059] S1: Based on the current state parameters of the water surface obstacle and environmental parameters, construct a target factor set and determine the hierarchical structure; based on the hierarchical structure, construct an expert experience model; based on the maximum deviation method, construct a mathematical model; construct a backpropagation neural network and output an intelligent model; integrate the expert experience model, the mathematical model, and the intelligent model to determine the collision risk assessment model and solve for the risk assessment degree.

[0060] S2: Read the current danger assessment of the water surface obstacle, the position parameters of the unmanned surface vessel, and the navigation status parameters of the unmanned surface vessel, preprocess them, and transmit them to the obstacle avoidance path generation module;

[0061] S3: Depending on the specific situation, automatically adjust the number of nodes and population size of the dung beetle optimization algorithm, or keep the default settings.

[0062] S4: Improve the dung beetle algorithm by using the Chebyshev chaotic strategy, back learning, simulated polynomial mutation, and Levy flight strategy to determine the dung beetle optimization algorithm; iterate the optimal solution of the dung beetle optimization algorithm according to the simplex method; generate the obstacle avoidance path of the unmanned surface vessel through the obstacle avoidance path generation module and the optimal solution.

[0063] Step S1 includes:

[0064] S11: Using the sensors of the unmanned surface vessel, read the state parameters and environmental parameters, and construct the target factor set M as follows:

[0065] M = [γ DCPA γTCPA γ D γ B γ K ]

[0066] Where, γ DCPA γ is the minimum encounter distance index between an unmanned surface vessel and an obstacle. TCPA γ is the minimum encounter time index between an unmanned surface vessel and an obstacle. D γ is an indicator of the distance between obstacles. B Azimuth difference index, γ K Speed ​​ratio index;

[0067] S12: Arrange the influence of the five indicators in the target factor set M in a hierarchical structure to determine the hierarchical structure;

[0068] Based on expert experience in assessing the risk level of unmanned surface vessels (USVs) during obstacle avoidance in geological survey operations, a judgment matrix is ​​determined and its consistency is checked. The eigenvector corresponding to the largest eigenvalue of the judgment matrix after the check is then calculated and normalized to obtain the weight vector J of the expert experience model. ω,1 ;

[0069] Based on the hierarchical structure and weight vector J ω,1 Construct the expert experience model CRI1:

[0070] CRI1=J ω,1 ·M

[0071] S13: Establish the attribute difference values ​​in the target factor set M, transform the weight allocation problem into an optimization problem, calculate the distance between each scheme and the positive and negative ideal solutions, and define the deviation of different attributes of each scheme from the corresponding positive and negative ideal solution components;

[0072] Establish the Lagrange function and solve for J. ω,2 ;

[0073] Based on the maximum deviation method, an optimization model is given according to the principle of maximizing deviation; weights are assigned to the five indicators in the target factor set M according to the attribute difference values; based on the optimization model and the weights of the five indicators, the mathematical model CRI2 is constructed as follows:

[0074] CRI2=J ω,2 ·M

[0075] S14: Construct the backpropagation neural network, and set the structure of the backpropagation neural network to 5~10~1, representing an input layer with 5 input nodes, a hidden layer with 10 nodes, and an output layer with 1 node;

[0076] The input of the backpropagation neural network is five indicators in the target factor set M;

[0077] The backpropagation neural network outputs the intelligent model CRI3 through forward learning and error backpropagation;

[0078] S15: Determine the weights of the collision risk assessment model. According to the weights, fuse the expert experience model, the mathematical model, and the intelligent model to determine the collision risk assessment model, and solve the risk assessment degree CRI.

[0079] Specifically, use the Lingo tool in Matlab to assign weights to the five indicators in the target factor set M.

[0080] Specifically, since the number of hidden units in the neural network is closely related to the number of layers, an increase in the number of layers will lead to the complication of the classification boundary mapping function, increase the training time, and cause overfitting; while a decrease in the number of hidden units will lead to a decrease in the convergence speed and cause underfitting. Through the error method, it is found that the average error is the smallest when the number of hidden nodes = 10. Therefore, the backpropagation neural network structure is selected. The learning algorithm of the backpropagation neural network uses the backpropagation of the network output error to iteratively update the weight parameters of the backpropagation neural network, and finally achieves the minimum error, including forward learning and error backpropagation. The output result of the backpropagation neural network is CRI3.

[0081] Step S2 includes:

[0082] According to the risk assessment degree CRI, divide the safety levels as follows:

[0083] If CRI ≥ 0.4, the safety level is "safe";

[0084] If 0.4 < CRI ≤ 0.6, the safety level is "threatened", and evasion should be considered;

[0085] If 0.6 < CRI ≤ 0.8, the safety level is "highly threatened", and evasion should be given priority;

[0086] If 0.8 < CRI ≤ 1.0, the safety level is "dangerous", and evasion should be carried out immediately, and the obstacle avoidance priority is higher than that of other obstacles;

[0087] After sorting the risk assessment degrees before the detected water surface obstacles, transmit them to the obstacle avoidance path generation module together with the position parameters and the navigation state parameters.

[0088] Step S3 includes:

[0089] The default settings of the dung beetle optimization algorithm are: the population size is set to 120, the number of iterations is 300, and the number of node insertions is 200;

[0090] Based on the number of obstacles and the aforementioned hazard assessment, the number of nodes and population size of the dung beetle optimization algorithm are adjusted as follows:

[0091] Case 1: When the number of obstacles is greater than or equal to 3 and the danger assessment score is greater than or equal to 0.8, increase the population size, decrease the number of iterations, and reduce the number of node insertions;

[0092] Case 2: When the number of obstacles is less than or equal to 1 and the danger assessment degree is less than 0.6, reduce the population size, reduce the number of iterations, and reduce the number of node insertions;

[0093] When the number of obstacles does not match the risk assessment value in cases one and two, the dung beetle optimization algorithm is adjusted to the default setting.

[0094] Step S4 includes:

[0095] S41: Initialize the dung beetle position using the Chebyshev chaotic strategy to improve the initialization settings of the dung beetle optimization algorithm;

[0096] S42: Improve the learning rate of the dung beetle optimization algorithm through the reverse learning;

[0097] S43: Perform a perturbation mutation operation at the optimal solution position of the dung beetle optimization algorithm using the Cauchy mutation operator;

[0098] S44: By using the Levy flight strategy, the randomness of the individual search step size in the dung beetle optimization algorithm is improved;

[0099] S45: The optimal solution of the dung beetle optimization algorithm is obtained iteratively using the simplex method; the obstacle avoidance path of the unmanned surface vessel is generated using the obstacle avoidance path generation module and the optimal solution.

[0100] Specifically, the dung beetle positions are initialized using a Chebyshev chaotic strategy, resulting in a wider and more comprehensive distribution of dung beetles. A reverse learning mechanism is used to obtain a reverse solution, expanding the search domain of the dung beetle optimization algorithm. A Cauchy mutation operator is used to perturb and mutate the optimal solution position of the dung beetle optimization algorithm, obtaining a new solution to improve its optimization performance and mitigate its tendency to get trapped in local optima. A Levy flight strategy is used to improve the randomness of individual search step lengths, enhancing the effectiveness of the population search range and global comprehensive search capability, ensuring information supply for superior individuals, and alternating the algorithm's movement pattern between frequent short-distance jumps and occasional long-distance jumps, facilitating escape from local optima. A simplex optimization algorithm is used to initialize basic variables, merge them with relaxation variables to form a matrix, determine if there are any elements that can be further increased, record the substituted variables in an array, and finally return the target value. The improved dung beetle optimization algorithm is run to iteratively optimize and generate obstacle avoidance path node positions, ultimately forming the unmanned surface vessel's obstacle avoidance path.

[0101] Specifically, Figure 3 This image shows a comparison of obstacle avoidance paths between the improved Dung Beetle Optimization Algorithm (IDBO) and other biomimetic algorithms (Dung Beetle Optimization Algorithm (DBO), Particle Swarm Optimization Algorithm (PSO), and Grey Wolf Optimization Algorithm (GWO). The improved Dung Beetle Optimization Algorithm iteratively generates obstacle avoidance paths for unmanned surface vessels. A comparison with other common intelligent biomimetic algorithms clearly shows that the improved Dung Beetle Optimization Algorithm offers a safer strategy and easier path planning when there are many obstacles, and a shorter and more accurate path when there are few obstacles.

[0102] Figure 4 The central pentagram represents the CRI value calculated by the software, while the black dot represents the CRI value predicted by the collision hazard assessment model.

[0103] This application selects five hazard assessment indicators for the navigation process; calculates the hazard assessment scores of the expert experience model, mathematical model, and intelligent model in sequence, and fuses them according to certain weights to obtain the obstacle hazard assessment score; sorts obstacles according to the value and range of the hazard assessment score and provides decision support for obstacle avoidance path generation, adjusting the initial settings of the obstacle avoidance algorithm. Finally, the improved dung beetle algorithm is used to solve for the obstacle avoidance path under the current situation.

[0104] Specifically, the process involves: 1. Information collection and processing. 2. Using the UAV's built-in radar and vision software, collecting current UAV status parameters, environmental parameters, and obstacle status information to calculate the minimum encounter distance, minimum encounter time, distance to the obstacle, bearing difference, and speed ratio. A total of five detection parameters are used. Obstacles are detected a total of 90 times during the navigation process, resulting in a sample size of 90. 3. Calculating the hazard assessment degree. 4. Adjusting the number of nodes, population size, and iteration count of the dung beetle optimization algorithm. Due to the large number of obstacles in the first half, the number of nodes is reduced, the population size is increased to 150, the iteration count is reduced to 220, and the number of inserted nodes is reduced to 180. In the second half, with a large number of obstacles, default parameters (population size set to 120, iteration count to 300, and number of inserted nodes to 200) are used for obstacle avoidance path planning. 5. Generating the UAV obstacle avoidance path.

[0105] Specifically, a large difference in attribute values ​​indicates a significant impact on the assessment.

[0106] It should be noted that the above embodiments of the apparatus are only illustrated by the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. In addition, the apparatus and method embodiments provided in the above embodiments belong to the same concept, and the specific implementation process can be found in the method embodiments, which will not be repeated here.

[0107] This application also discloses an electronic device. (See reference...) Figure 5 , Figure 5 This is a schematic diagram of the structure of an electronic device disclosed in an embodiment of this application. The electronic device 500 may include: at least one processor 501, at least one network interface 504, a user interface 503, a memory 505, and at least one communication bus 502.

[0108] The communication bus 502 is used to enable communication between these components.

[0109] The user interface 503 may include a display screen and a camera. Optionally, the user interface 503 may also include a standard wired interface and a wireless interface.

[0110] The network interface 504 may optionally include a standard wired interface or a wireless interface (such as a Wi-Fi interface).

[0111] The processor 501 may include one or more processing cores. The processor 501 connects to various parts of the server using various interfaces and lines, and performs various server functions and processes data by running or executing instructions, programs, code sets, or instruction sets stored in memory 505, and by calling data stored in memory 505. Optionally, the processor 501 may be implemented using at least one hardware form of Digital Signal Processing (DSP), Field-Programmable Gate Array (FPGA), or Programmable Logic Array (PLA). The processor 501 may integrate one or a combination of several of the following: Central Processing Unit (CPU), Graphics Processing Unit (GPU), and modem. The CPU primarily handles the operating system, user interface, and applications; the GPU is responsible for rendering and drawing the content required for display; and the modem handles wireless communication. It is understood that the modem may also not be integrated into the processor 501 and may be implemented as a separate chip.

[0112] The memory 505 may include random access memory (RAM) or read-only memory.

[0113] Optionally, the memory 505 includes a non-transitory computer-readable storage medium. The memory 505 can be used to store instructions, programs, code, code sets, or instruction sets. The memory 505 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as touch functionality, sound playback functionality, image playback functionality, etc.), instructions for implementing the various method embodiments described above, etc.; the data storage area may store data involved in the various method embodiments described above, etc. The memory 505 may also optionally include, but is not limited to, at least one storage device located remotely from the aforementioned processor 501. (Refer to...) Figure 5 The memory 505, which serves as a computer storage medium, may include an operating system, a network communication module, a user interface module, and an application program for generating obstacle avoidance paths for unmanned surface vessels using an improved dung beetle optimization algorithm.

[0114] exist Figure 5In the illustrated electronic device 500, the user interface 503 is mainly used to provide an input interface for the user and to acquire user input data; while the processor 501 can be used to call the application program stored in the memory 505, which is an improved dung beetle optimization algorithm for generating obstacle avoidance paths for unmanned surface vessels. When executed by one or more processors 501, the electronic device 500 performs one or more methods as described in the above embodiments. It should be noted that, for the foregoing method embodiments, for the sake of simplicity, they are all described as a series of actions. However, those skilled in the art should understand that this application is not limited to the described order of actions, because according to this application, some steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also understand that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily essential to this application.

[0115] In the above embodiments, the descriptions of each embodiment have different focuses. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions in other embodiments.

[0116] In the various embodiments provided in this application, it should be understood that the disclosed apparatus can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the couplings or direct couplings or communication connections shown or discussed include, but are not limited to, indirect couplings or communication connections between apparatuses or units through some service interfaces, including but not limited to electrical or other forms.

[0117] The units described as separate components include, but are not limited to, physically separate units. Components shown as units include, but are not limited to, physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of these units can be selected to achieve the purpose of this embodiment according to actual needs.

[0118] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or may exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0119] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage device (CMD). Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a memory and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this application. The aforementioned memory includes various media capable of storing program code, such as USB flash drives, portable hard drives, magnetic disks, or optical disks.

[0120] The above are merely exemplary embodiments of this disclosure and should not be construed as limiting the scope of this disclosure. Any equivalent changes and modifications made in accordance with the teachings of this disclosure shall still fall within the scope of this disclosure. Other embodiments of this disclosure will readily conceive of those skilled in the art upon consideration of the specification and the disclosure of practical truths.

[0121] This application is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not described in this disclosure. The specification and embodiments are to be considered exemplary only, and the scope and spirit of this disclosure are defined by the claims.

Claims

1. A method for generating obstacle avoidance paths for unmanned surface vessels based on an improved dung beetle optimization algorithm, characterized in that, The method includes the following steps: S1: Based on the current state parameters of the water surface obstacles and environmental parameters, construct a target factor set and determine the hierarchical structure; based on the hierarchical structure, construct an expert experience model; based on the maximum deviation method, construct a mathematical model; construct a backpropagation neural network and output an intelligent model; By integrating the expert experience model, the mathematical model, and the intelligent model, a collision risk assessment model is determined, and the risk assessment degree is solved. Step S1 includes: S11: Using the sensors of the unmanned surface vessel, read the state parameters and environmental parameters, and construct the target factor set M as follows: in, This is the minimum encounter distance indicator between unmanned surface vessels and obstacles. This is the minimum encounter time indicator between the unmanned surface vessel and obstacles. This is an indicator of the distance between obstacles. Azimuth difference index Speed ​​ratio index; S12: For the target factor set The influence of the five indicators in the data is arranged hierarchically to determine the hierarchical structure. Based on expert experience in assessing the risk level of unmanned surface vessels (USVs) during obstacle avoidance in geological survey operations, a judgment matrix is ​​determined and its consistency is checked. The eigenvector corresponding to the largest eigenvalue of the judgment matrix after the check is then calculated and normalized to obtain the weight vector of the expert experience model. ; Based on the hierarchical structure and weight vector Build expert experience model : S13: Establish the target factor set The attribute difference values ​​in the data transform the weight allocation problem into an optimization problem. The distance between each scheme and the positive and negative ideal solutions is calculated, and the deviation of different attributes of each scheme from the corresponding positive and negative ideal solution components is defined. Establish the Lagrange function and solve for it. ; Based on the maximum deviation method and the principle of maximizing deviation, an optimization model is given; the target factor set is then analyzed based on the attribute difference values. The five indicators are assigned weights; based on the optimization model and the weights of the five indicators, a mathematical model is constructed. ,as follows: S14: Construct the backpropagation neural network, and set the structure of the backpropagation neural network to 5~10~1, representing an input layer with 5 input nodes, a hidden layer with 10 nodes, and an output layer with 1 node; The input to the backpropagation neural network is five indicators from the target factor set M; The backpropagation neural network outputs an intelligent model through forward learning and error backpropagation. ; S15: Determine the weights of the collision hazard assessment model; based on the weights, integrate the expert experience model, the mathematical model, and the intelligent model to determine the collision hazard assessment model, and solve for the hazard assessment degree. ; S2: Read the current danger assessment of the water surface obstacle, the position parameters of the unmanned surface vessel, and the navigation status parameters of the unmanned surface vessel, preprocess them, and transmit them to the obstacle avoidance path generation module; S3: Depending on the specific situation, automatically adjust the number of nodes and population size of the dung beetle optimization algorithm, or keep the default settings. S4: Improve the dung beetle algorithm by using the Chebyshev chaotic strategy, back learning, simulated polynomial mutation, and Levy flight strategy to determine the dung beetle optimization algorithm; iterate the optimal solution of the dung beetle optimization algorithm according to the simplex method; generate the obstacle avoidance path of the unmanned surface vessel through the obstacle avoidance path generation module and the optimal solution; Step S4 includes: S41: Initialize the dung beetle position using the Chebyshev chaotic strategy to improve the initialization settings of the dung beetle optimization algorithm; S42: Improve the learning rate of the dung beetle optimization algorithm through the reverse learning; S43: Perform a perturbation mutation operation at the optimal solution position of the dung beetle optimization algorithm using the Cauchy mutation operator; S44: By using the Levy flight strategy, the randomness of the individual search step size in the dung beetle optimization algorithm is improved; S45: The optimal solution of the dung beetle optimization algorithm is obtained iteratively using the simplex method; the obstacle avoidance path of the unmanned surface vessel is generated using the obstacle avoidance path generation module and the optimal solution.

2. The unmanned surface vessel obstacle avoidance path generation method based on the improved dung beetle optimization algorithm as described in claim 1, characterized in that, Step S2 includes: Based on risk assessment The security levels are classified as follows: like If so, the security level is "safe"; like If so, the safety level is "threatening" and you should consider taking shelter. like If the safety level is "highly dangerous", then avoidance should be the priority. like If it is a dangerous obstacle, the safety level is "dangerous" and you should take immediate evasive action. The obstacle avoidance priority is higher than that of other obstacles. After sorting the danger assessment scores of the detected water surface obstacles, the scores are transmitted to the obstacle avoidance path generation module along with the position parameters and the navigation status parameters.

3. The unmanned surface vessel obstacle avoidance path generation method based on the improved dung beetle optimization algorithm as described in claim 1, characterized in that, Step S3 includes: The default settings for the dung beetle optimization algorithm are: population size of 120, number of iterations of 300, and number of node insertions of 200. Based on the number of obstacles and the aforementioned hazard assessment, the number of nodes and population size of the dung beetle optimization algorithm are adjusted as follows: Case 1: When the number of obstacles is greater than or equal to 3 and the danger assessment score is greater than or equal to 0.8, increase the population size, decrease the number of iterations, and reduce the number of node insertions; Case 2: When the number of obstacles is less than or equal to 1 and the danger assessment degree is less than 0.6, reduce the population size, reduce the number of iterations, and reduce the number of node insertions; When the number of obstacles does not match the risk assessment value in cases one and two, the dung beetle optimization algorithm is adjusted to the default setting.

4. An electronic device, characterized in that, The device includes a processor (501), a memory (505), a user interface (503), and a network interface (504). The memory (505) is used to store instructions. The user interface (503) and the network interface (504) are used to communicate with other devices. The processor (501) is used to execute the instructions stored in the memory (505) to cause the electronic device (500) to perform the method as described in any one of claims 1-3.

5. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores instructions that, when executed, perform the steps of the method as described in any one of claims 1-3.

Images