Multi-unmanned ship task allocation method based on starfish behaviors

Abstract

The invention discloses a multi-unmanned-ship task allocation method based on starfish behaviors, and the method is suitable for solving the task allocation problem of multiple unmanned ships in cooperative operation, and especially has important application in the aspect of cooperative obstacle breaking of non-cooperative targets. The method comprises the following steps: firstly, randomly generating a position matrix of each starfish in an initialization stage and calculating fitness, then updating the positions of the starfishes by adopting different search modes according to the dimension of a problem, ensuring that the updated positions of the starfishes are within the boundary of a design variable, and updating the positions of the starfishes in predation and regeneration stages by utilizing a parallel bidirectional search strategy, and finally, outputting a globally optimal solution, and applying an optimization result to the cooperative obstacle breaking task of the unmanned surface vehicle, thereby improving the task execution efficiency and cooperative combat ability of the unmanned surface vehicle in a complex environment.

Description

Technical Field

[0001] The present invention belongs to the technology of unmanned cluster cooperative control, and is applicable to the multi-unmanned boat cooperative obstacle-breaking task. In particular, it relates to a multi-unmanned boat task allocation method based on starfish behavior. Background Art

[0002] In recent years, swarm intelligence optimization methods have been widely used in solving various complex problems. Especially in situations lacking gradient information, these methods have shown excellent performance. Despite significant progress in theory and practical applications, they still face some challenges, such as slow convergence speed, being prone to falling into local optimal solutions, high algorithm complexity, and poor stability in some practical problems. These problems have prompted researchers to continuously explore new optimization algorithms. Taking the task allocation of unmanned boats as an example, solving this optimization problem not only requires dealing with complex marine environments but also considering multiple factors such as obstacle avoidance and limited energy constraints.

[0003] In this context, the starfish optimization algorithm is proposed. Its inspiration comes from the predation, regeneration, and exploration behaviors of starfish. For problems with higher dimensions, a five-dimensional search mode is adopted; for problems with lower dimensions, a one-dimensional search strategy is used to flexibly meet the dimensional requirements of different optimization problems. The predation and regeneration stages are carried out in parallel, which improves the convergence speed and global search ability of the algorithm and helps to avoid local optimal solutions. Compared with traditional optimization algorithms, the biggest feature of this multi-unmanned boat task allocation method based on starfish behavior is that it not only improves the efficiency of task allocation but also can maintain high stability and robustness in complex and dynamically changing environments, providing a powerful optimization tool for the autonomous cooperation of unmanned boat swarms. Summary of the Invention

[0004] The present invention provides a multi-unmanned boat task allocation method based on starfish behavior to solve the problems existing in the above-mentioned prior art.

[0005] To achieve the above object, the present invention provides a multi-unmanned boat task allocation method based on starfish behavior, and the method includes the following steps:

[0006] S1. In the initialization stage, randomly generate the position matrix of each starfish and calculate the fitness;

[0007] S2. According to different dimensional limitations, update the positions of starfish using different search modes. If the updated position of a starfish exceeds the boundary of the design variables, keep its position at the original position;

[0008] S3. Adopt a parallel bidirectional search strategy to update the positions of starfish in the predation stage and the regeneration stage;

[0009] S4. Repeat the position update process until the maximum number of iterations is reached, output the global optimal solution, and use the result of task allocation for the collaborative obstacle breaking of the unmanned surface vehicle.

[0010] Preferably, the specific content of S1 is as follows:

[0011] In the initialization stage, randomly generate the position matrix of each starfish and calculate the fitness. The position of the starfish is randomly generated between the boundaries of the design variables and can be represented as a matrix:

[0012]

[0013] where X is the matrix used to save the position of the starfish, with a size of N×D, N being the population size, and D being the dimension of the design variable. In the initialization stage, the position of each starfish is:

[0014] X ij = l j + r(u j - l j ) i = 1, 2,..., N j = 1, 2,..., D

[0015] where X ij represents the position information of the i-th starfish in the j-th dimension, r is a random number between (0, 1), u j and l j are respectively the upper and lower bounds of the j-th dimension design variable, N is the population size, and D is the dimension of the design variable.

[0016] After generating the initialization position matrix, obtain the fitness values of all starfish by evaluating the objective function, and these fitness values can also be stored as a vector:

[0017]

[0018] where F is a matrix used to store and update the obtained fitness values, with a size of N×1, N being the population size.

[0019] Preferably, the specific content of S2 is as follows:

[0020] According to different dimension restrictions, update the position of the starfish using different search patterns. If the updated position of the starfish exceeds the boundaries of the design variable, keep its position at the original position.

[0021] If the dimension D of the optimization problem is greater than 5, the search space of this problem is large, and it is required that the starfish move all five tentacles to explore the surrounding environment. The mathematical model of this stage is:

[0022]

[0023] where T is the current iteration number, and represent the p - th dimension of the position obtained by the i - th starfish and the current position respectively, represents the p - th dimension of the current best position, p is one of the five dimensions randomly selected from D dimensions, r is a random number between (0, 1), and the calculation formulas of parameters a1 and θ are as follows:

[0024] a1=(2r - 1)π

[0025]

[0026] where T is the current iteration number, T max is the maximum iteration number, r is a random number between (0, 1). The sine and cosine terms indicate that the tentacles of the starfish may twist left or right with the same probability to approach the food. In the exploration stage, a1 is randomly generated in each candidate solution and iteration to update the position, and θ changes with the increase of the iteration number. These two parameters can measure the distance effect between the best position and the currently selected updated dimension.

[0027] For optimization problems with dimension D less than 5, a five - dimensional search pattern is used, and only five dimensions of the starfish position are updated. When the updated position exceeds the boundary of the design variables, the tentacles of the starfish remain at the previous position instead of moving to the updated position, which is expressed by the mathematical formula as:

[0028]

[0029] where T is the current iteration number, and represent the p - th dimension of the position obtained by the i - th starfish and the current position respectively, p represents the updated dimension, l b，p and u b，p represent the lower and upper bounds of the design variables respectively.

[0030] If the dimension D of the optimization problem is not greater than 5, a one - dimensional search pattern is used in the exploration stage to update the position. The updated position can be expressed as:

[0031]

[0032] where T is the current iteration number, and represent the p - th dimension of the position obtained by the i - th starfish and the current position respectively, and are the p - th dimension positions of two randomly selected starfishes respectively, A1 and A2 are two random numbers between (-1, 1), p is a dimension randomly selected from D dimensions, E tIt is the energy of the starfish, and the calculation formula is:

[0033]

[0034] where T is the current iteration number, and T max is the maximum iteration number, and E t is the energy of the starfish.

[0035] Preferably, the specific content of S3 is:

[0036] Adopt a parallel bidirectional search strategy to update the position of the starfish during the predation stage and the regeneration stage;

[0037] Calculate five distances between the current best position and other starfish, and then randomly select two distances as references for updating the position of the starfish, and use the parallel bidirectional search strategy for updating. These distances can be calculated by the following formula:

[0038]

[0039] where d m are the five distances between the current global best position and other starfish, represents the current best position, m p are five randomly selected starfish, represents the positions of five randomly selected starfish, and T is the current iteration number. Therefore, the update rule of each starfish during the predation behavior can be expressed as:

[0040]

[0041] where T is the current iteration number, Y i T and respectively represent the position obtained by the i-th starfish and the current position, r1 and r2 are random numbers between (0, 1), and d m1 and d m2 are the distances randomly selected from d m Based on the parallel bidirectional search strategy, the candidate solutions of the starfish move towards better guiding solutions, while other candidate solutions move backward in the same iteration, and these candidate solutions have the same ability to overcome local optimal solutions;

[0042] The regeneration stage is only implemented in the last starfish (i = N) in the population, and the position is updated by the following formula:

[0043]

[0044] where T is the current iteration number, and T max represents the maximum iteration number, N is the size of the population, and Y iT and respectively represent the position obtained by the $i$-th starfish and the current position.

[0045] Preferably, the specific content of S4 is as follows:

[0046] Repeat the position update process until the maximum number of iterations is reached, output the global optimal solution, and use the result of task allocation for the collaborative obstacle-breaking of surface unmanned boats.

[0047] Implementing the embodiments of the present invention will have the following beneficial effects:

[0048] A multi-unmanned-boat task allocation method based on starfish behavior proposed by the present invention can flexibly search in the optimization space of different dimensions by simulating the predation, regeneration, and exploration behaviors of starfish, thereby avoiding the common local optimal solution problem in traditional methods. This mechanism enables the algorithm to improve the efficiency of task allocation, reduce the computational amount, and can also adaptively adjust the search strategy in a dynamically changing environment, thus ensuring the smooth completion of the collaborative obstacle-breaking task of unmanned boats. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] To more clearly illustrate the embodiments of the present invention, the related drawings are introduced below. The drawings show schematic diagrams of certain embodiments of the present invention, aiming to provide sufficient reference for technicians so that they can obtain more relevant information without the need for creative thinking. Through these drawings, the structure, components, and operation process of the present invention can be intuitively understood, thereby better understanding the specific details of the embodiments. The content of the drawings is as follows:

[0050] Figure 1 is a flowchart of a multi-unmanned-boat task allocation method based on starfish behavior;

[0051] Figure 2 is the multi-unmanned-boat task allocation result in one embodiment. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0052] The technical solutions will be described in detail and clearly below in conjunction with the drawings in the embodiments of the present invention. It should be emphasized that the embodiments are only a part of the demonstrations of the present invention and do not limit the only form of all embodiments. Based on the embodiments of the present invention, any situation that can be deduced by those skilled in the art without creative work should be regarded as a part of the protection scope of the present invention. Implementing the embodiments of the present invention will bring the following significant advantages: To solve the deficiencies in the prior art, this embodiment proposes a multi-unmanned-boat task allocation method based on starfish behavior, as Figures 1-2 shown, and the specific method includes:

[0053] S1. Initialization stage: Randomly generate the position matrix of each starfish and calculate the fitness; S2. According to different dimensional constraints, use different search patterns to update the positions of the starfish. If the updated position of a starfish exceeds the boundaries of the design variables, keep its position at the original position; S3. Adopt a parallel bidirectional search strategy to update the positions of the starfish during the predation stage and the regeneration stage; S4. Repeat the position update process until the maximum number of iterations is reached, output the global optimal solution, and use the result of the task allocation for the collaborative obstacle breaking of the surface unmanned boat.

[0054] In some specific embodiments, the specific content of S1 is as follows:

[0055] In the initialization stage, randomly generate the position matrix of each starfish and calculate the fitness. The position of the starfish is randomly generated between the boundaries of the design variables and can be represented as a matrix:

[0056]

[0057] where X is a matrix used to store the positions of the starfish, with a size of N×D, N being the population size, and D being the dimension of the design variables. In the initialization stage, the position of each starfish is:

[0058] X ij = l j + r(u j - l j ) i = 1, 2,..., N j = 1, 2,..., D

[0059] where X ij represents the position information of the i-th starfish in the j-th dimension, r is a random number between (0, 1), u j and l j are respectively the upper and lower bounds of the j-th dimensional design variable, N is the population size, and D is the dimension of the design variables.

[0060] After generating the initialization position matrix, obtain the fitness values of all starfish by evaluating the objective function, and these fitness values can also be stored as a vector:

[0061]

[0062] where F is a matrix used to store and update the obtained fitness values, with a size of N×1, N being the population size.

[0063] In some specific embodiments, the specific content of S2 is as follows:

[0064] According to different dimensional constraints, use different search patterns to update the positions of the starfish. If the updated position of a starfish exceeds the boundaries of the design variables, keep its position at the original position.

[0065] If the dimension D of the optimization problem is greater than 5, the search space of the problem is large, and it is required that the starfish move all five tentacles to explore the surrounding environment. The mathematical model at this stage is as follows:

[0066]

[0067] where T is the current iteration number, and represent the position obtained by the i-th starfish and the p-th dimension of the current position respectively, represents the p-th dimension of the current best position. p is one of the five dimensions randomly selected from the D dimensions. r is a random number between (0, 1). The calculation formulas for the parameters a1 and θ are as follows:

[0068] a1 = (2r - 1)π

[0069]

[0070] where T is the current iteration number, T max is the maximum iteration number, and r is a random number between (0, 1). The sine and cosine terms indicate that the tentacles of the starfish may twist left or right with the same probability to approach the food. In the exploration stage, a1 is randomly generated in each candidate solution and iteration to update the position, and θ changes with the increase of the iteration number. These two parameters can measure the distance effect between the best position and the currently selected updated dimension.

[0071] For optimization problems with dimension D less than 5, a five-dimensional search mode is used, and only five dimensions of the starfish position are updated. When the updated position exceeds the boundary of the design variables, the tentacles of the starfish remain at the previous position instead of moving to the updated position. It is expressed by the mathematical formula as:

[0072]

[0073] where T is the current iteration number, and represent the position obtained by the i-th starfish and the p-th dimension of the current position respectively. p represents the updated dimension, l b，p and u b，p represent the lower bound and upper bound of the design variables respectively.

[0074] If the dimension D of the optimization problem is not greater than 5, a one-dimensional search mode is used to update the position in the exploration stage. The updated position can be expressed as:

[0075]

[0076] where T is the current iteration number, and respectively represent the p - dimension of the position obtained by the i - th starfish and the current position, and are respectively the p - dimension positions of two randomly selected starfishes, A1 and A2 are two random numbers between (-1, 1), p is a dimension randomly selected from the D dimensions, E t is the energy of the starfish, and the calculation formula is:

[0077]

[0078] where T is the current iteration number, T max is the maximum iteration number, E t is the energy of the starfish.

[0079] In some specific embodiments, the specific content of S3 is:

[0080] Adopt a parallel bidirectional search strategy to update the positions of starfishes in the predation stage and the regeneration stage;

[0081] Calculate five distances between the current best position and other starfishes, then randomly select two distances as references for updating the positions of starfishes, and use the parallel bidirectional search strategy for updating. These distances can be calculated by the following formula:

[0082]

[0083] where d m are five distances between the current global best position and other starfishes, represents the current best position, m p are five randomly selected starfishes, represents the positions of five randomly selected starfishes, T is the current iteration number. Therefore, the update rule of each starfish in the predation behavior can be expressed as:

[0084]

[0085] where T is the current iteration number, Y i T and respectively represent the position obtained by the i - th starfish and the current position, r1 and r2 are random numbers between (0, 1), d m1 and d m2 are distances randomly selected from d m . Based on the parallel bidirectional search strategy, the candidate solutions of starfishes move towards better guiding solutions, while other candidate solutions move backward in the same iteration, and these candidate solutions have the same ability to overcome local optimal solutions;

[0086] The regeneration stage is only implemented in the last starfish in the population (i = N), and the position is updated by the following formula:

[0087]

[0088] where T is the current iteration number, T max represents the maximum iteration number, N is the size of the population, and Y i T and represent the position obtained by the i-th starfish and the current position, respectively.

[0089] In some specific embodiments, the specific content of S4 is as follows:

[0090] Suppose there are 4 unmanned boats in the marine environment, and they need to complete 10 tasks to be attacked. Suppose these tasks are distributed in a sea area of 100km × 100km. Each task has different importance, and each unmanned boat has its fixed starting position. The goal of the task is to achieve the overall optimal attack effect by reasonably allocating tasks to the unmanned boats, repeat the position update process until the maximum iteration number is reached, output the global optimal solution, and use the result of the task allocation for the collaborative obstacle breaking of the surface unmanned boats to obtain the result of the multi-unmanned boat task allocation.

[0091] The present invention aims at the problem of multi-unmanned boat task allocation in the existing algorithm, and proposes a multi-unmanned boat task allocation method based on starfish behavior. By simulating the starfish behavior, this method can not only effectively explore the task space, but also avoid falling into the local optimal solution by introducing parallel bidirectional search strategies, five-dimensional and one-dimensional search modes, etc. Specifically, this method makes the task allocation process more flexible and efficient by simulating the predation behavior and regeneration behavior of starfish.

[0092] The above content is only an application example of the present invention in the field of multi-unmanned boat task allocation, and does not limit its application scope. The actual protection scope includes more implementation manners and technical solutions. Any improvement, change or other implementation manner based on the principle of the present invention shall be regarded as a part of the present invention as long as it does not deviate from its core idea. Therefore, the protection scope of the present invention allows reasonable adjustment and innovation without violating the basic principle.

Claims

1. A multi-unmanned boat mission allocation method based on starfish behavior, characterized in that It includes the following steps: S1. Initialization phase: Randomly generate the position matrix of each starfish and calculate the fitness; S2. According to different dimensionality constraints, use different search patterns to update the positions of the starfish. If the updated position of a starfish exceeds the boundary of the design variables, its position is maintained at the original position; S3. Adopt a parallel bidirectional search strategy to update the positions of the starfish during the predation phase and the regeneration phase; S4. Repeat the position update process until the maximum number of iterations is reached, output the global optimal solution, and use the result of task assignment for the cooperative obstacle breaking of the unmanned surface vehicle.

2. The multi-unmanned boat mission allocation method based on starfish behavior according to claim 1, characterized in that: The specific content of S1 is: In the initialization phase, randomly generate the position matrix of each starfish and calculate the fitness; In the initialization phase, the positions of the starfish are randomly generated between the boundaries of the design variables and can be represented as a matrix: where X is the matrix used to store the positions of the starfish, with a size of N×D, N being the population size, and D being the dimensionality of the design variables. In the initialization phase, the position of each starfish is: X ij = l j + r(u j - l j ) i = 1, 2, ..., N j = 1, 2, ..., D Among them, X ij represents the position information of the i-th starfish in the j-th dimension, r is a random number between (0, 1), u j and l j are the upper and lower bounds of the j-th design variable respectively, N is the population size, and D is the dimension of the design variable; After generating the initialization position matrix, obtain the fitness values of all starfish by evaluating the objective function, and these fitness values can also be stored as a vector: where F is a matrix used to store and update the obtained fitness values, with a size of N×1, N being the population size.

3. A multi-unmanned boat mission allocation method based on starfish behavior according to claim 1, characterized in that: The specific content of S2 is: According to different dimensionality constraints, use different search patterns to update the positions of the starfish. If the updated position of a starfish exceeds the boundary of the design variables, its position is maintained at the original position; If the dimensionality D of the optimization problem is greater than 5, the search space of this problem is large, and it is required that the starfish move all five tentacles to explore the surrounding environment. The mathematical model of this phase is: where T is the current iteration number, and respectively represent the p-th dimension of the position obtained by the i-th starfish and the current position, represents the p-th dimension of the current best position, p is one of the five dimensions randomly selected from D dimensions, r is a random number between (0, 1), and the calculation formulas for parameters a1 and θ are as follows: a1＝(2r-1)π where T is the current iteration number, and T max is the maximum iteration number, and r is a random number between (0, 1). The sine and cosine terms indicate that the tentacles of the starfish may twist left or right with the same probability to approach the food. In the exploration stage, a1 is randomly generated in each candidate solution and iteration to update the position, and θ changes with the increase of the iteration number. These two parameters can measure the distance effect between the optimal position and the currently selected updated dimension; For optimization problems with a dimensionality D less than 5, use a five-dimensional search pattern to only update five dimensions of the starfish's position. When the updated position exceeds the boundary of the design variables, the tentacles of the starfish remain at the previous position instead of moving to the updated position, which is expressed by the mathematical formula as: where T is the current iteration number, and represent the p-th dimension of the position obtained by the i-th starfish and the current position respectively, p represents the updated dimension, l b，p and u b，p represent the lower and upper bounds of the design variables respectively; If the dimensionality D of the optimization problem is not greater than 5, the one-dimensional search pattern is used to update the position during the exploration phase. The updated position can be expressed as: where T is the current iteration number, and represent the p-th dimension of the position obtained by the i-th starfish and the current position respectively, and are the p-th dimensional positions of two randomly selected starfish respectively. A1 and A2 are two random numbers between (-1, 1), and p is a dimension randomly selected from the D dimensions. E t is the energy of the starfish, and the calculation formula is: Among them, T is the current iteration number, and T max is the maximum iteration number, and E t is the energy of the starfish.

4. A multi-unmanned boat mission allocation method based on starfish behavior according to claim 1, characterized in that: The specific content of S3 is: Adopt a parallel bidirectional search strategy to update the positions of the starfish during the predation phase and the regeneration phase; Calculate the five distances between the current best position and other starfish, then randomly select two distances as references for updating the positions of the starfish, and use the parallel bidirectional search strategy for updating. These distances can be calculated by the following formula: Among them, d m are the five distances between the current global best position and other starfish, represents the current best position, m p are five randomly selected starfish, represents the positions of five randomly selected starfish, and T is the current iteration number. Therefore, the update rule for each starfish in the foraging behavior can be expressed as: where T is the current iteration number, and represent the position obtained by the i-th starfish and the current position respectively, r1 and r2 are random numbers between (0, 1), d m1 and d m2 are randomly selected distances from d m Based on the parallel bidirectional search strategy, the candidate solutions of the starfish move towards better guiding solutions, while other candidate solutions move backward in the same iteration, and these candidate solutions have the same ability to overcome local optima; The regeneration phase is only implemented in the last starfish (i = N) in the population, and the position is updated by the following formula: where T is the current iteration number, T max represents the maximum iteration number, N is the size of the population, and represent the position obtained by the i-th starfish and the current position, respectively.

5. A multi-unmanned boat mission allocation method based on starfish behavior according to claim 1, characterized in that: The specific content of S4 is: Repeat the position update process until the maximum number of iterations is reached, output the global optimal solution, and use the result of task assignment for the cooperative obstacle breaking of the unmanned surface vehicle.

Images