Task allocation and time collaboration method for multiple unmanned systems

Abstract

A task allocation and time collaboration method for multiple unmanned systems is characterized by comprising the following steps: (1) calculating a capability matching degree based on a function demand condition of a task needing to be executed in combination with state information of an unmanned platform; (2) defining a particle position vector to represent a task allocation strategy; (3) based on the capability matching degree between the unmanned platform and the tasks, constructing a multi-target fitness function corresponding to the particle position vector to represent the overall fitness degree when all the tasks are allocated according to the particle position vector; (4) constructing an initial population; iteration of an initial population is carried out on the basis of a multi-target fitness function through a genetic particle swarm algorithm, and an optimal particle position vector is obtained; and (5) distributing the tasks to the corresponding unmanned platforms according to the optimal particle position vectors to complete task distribution. Through the process, various factors such as performance parameters of multiple unmanned platforms and comprehensive efficiency during task execution are fully considered in the process of optimizing the task allocation scheme.

Description

Technical Field

[0001] This invention relates to a method for task allocation and time coordination in a multi-unmanned system, belonging to the field of unmanned system cooperative control technology. Background Technology

[0002] With the continuous development of unmanned systems technology, the collaborative execution of tasks by multiple types of unmanned platforms, such as drones, unmanned surface vessels (USVs), and unmanned ships, has become an important development trend. However, existing methods for task allocation among multiple unmanned platforms mainly target single types of unmanned platforms, making it difficult to effectively coordinate and allocate different types of unmanned platforms for task execution based on their performance parameters. Furthermore, existing methods typically only consider a single optimization objective when allocating tasks, failing to comprehensively consider factors such as the overall efficiency and load balancing of different unmanned platforms during task execution. Summary of the Invention

[0003] The technical problem solved by this invention is to overcome the shortcomings of existing technologies and provide a method for task allocation and time coordination in multi-unmanned systems. This method constructs a multi-objective fitness function and uses a genetic particle swarm optimization algorithm to iteratively optimize the task allocation scheme for multiple unmanned platforms. During the optimization process, it fully considers various factors such as the performance parameters of the multiple unmanned platforms and the overall efficiency during task execution, thus solving the problems of existing methods that can only allocate tasks to a single type of unmanned platform and only consider a single optimization objective during task allocation.

[0004] The technical solution of this invention is: A method for task allocation and time coordination in a multi-unmanned system, comprising the following steps: (1) Collection M Status information of heterogeneous unmanned platforms; based on the execution requirements. N Based on the functional requirements of each task and the status information of the unmanned platform, calculate the capability matching degree between each unmanned platform and each task; (2) Define the particle position vector ,in All are arbitrary integers between 1 and M; the particle position vector X Each element in the table corresponds to a task allocation strategy. Indicates the first i The task is assigned to the first An unmanned platform; (3) Based on the capability matching degree between the unmanned platform and the task, a multi-objective fitness function corresponding to the particle position vector is constructed; the calculation result of the multi-objective fitness function represents the overall fitness degree when all tasks are allocated according to the allocation strategy represented by the particle position vector. (4) Randomly generate multiple particle position vectors, and all particle position vectors constitute the initial population.Z t The initial population is determined using a genetic particle swarm optimization algorithm based on a multi-objective fitness function. Z t Through iteration, the optimal particle position vector is obtained. ; (5) Assign tasks to the corresponding unmanned platforms according to the allocation strategy corresponding to each element in the optimal particle position vector, and complete the task allocation.

[0005] Furthermore, the status information of the unmanned platform includes payload capacity, maximum speed, detection range, and communication radius.

[0006] Furthermore, the process for calculating the capability matching degree between the unmanned platform and the task in step (1) is as follows: (1.1) Based on the state information of the unmanned platform, extract the state information of each unmanned platform. L One performance parameter; (1.2) Each unmanned platform L The performance parameters are arranged in a preset performance parameter order to form the capability vector of the corresponding unmanned platform; among them, the first... i The capability vector of each unmanned platform is , For the first i The first unmanned platform k One performance parameter; (1.3) Based on the functional requirements of each task, statistically analyze the impact of each task on the UAV. L The requirements for each performance parameter are analyzed to construct a requirement vector for each task; where the first parameter is... j The demand vector for each task is: , For the first j The first task for the unmanned platform k Minimum required threshold for each performance parameter; (1.4) Based on the capability vector of each unmanned platform and the requirement vector of the task, calculate the capability matching degree between each unmanned platform and each task. The calculation formula is as follows:

[0007] in, For the first i The unmanned platform and the first j The degree of capability matching between tasks For the first k The weights of each performance parameter are given, and min(.) is the function to find the minimum value.

[0008] Furthermore, the specific process of constructing the multi-objective fitness function corresponding to the particle position vector in step (3) is as follows: (3.1) Based on the capability matching degree between the unmanned platform and the task, calculate the task benefit term when all tasks are allocated according to the particle position vector. The calculation formula is:

[0009] in, Indicates the first j Whether the first task has been assigned, when the first task is... j One task was assigned. If it is 1, when the first j One task has not been assigned. =0; For the first j The first task and the first The compatibility of capabilities between different task platforms; (3.2) Calculate the execution cost term when all tasks are assigned according to the particle position vector. The calculation formula is:

[0010] in, and All are weighting coefficients; For the first i The number of tasks assigned to each unmanned platform; For the first i The unmanned platform from the first j The task execution location is transposed to the first... j +Time required to execute 1 task position; For the first i The first unmanned platform to execute the j The processing time required for each task; (3.3) Calculate the time co-operation factor when all tasks are assigned according to the particle position vector. ; (3.4) Calculate the balance factor when all tasks are assigned according to the particle position vector. ; (3.5) Based on task benefit items Execution cost item Time co-factor and balance factor Calculate the particle position vector X Corresponding multi-objective fitness function F The formula is:

[0011] in, , , and These are the task reward items. Execution cost item Time co-factor and balance factor The corresponding weights; For task reward items The theoretical maximum value; For execution cost item The theoretical maximum value.

[0012] Furthermore, the time coordination factor in step (3.3) The calculation process is as follows: (3.3.1) Calculate the actual start time of each task after it has been assigned to the corresponding unmanned platform according to the particle position vector. The calculation formula is as follows:

[0013] in, For the first j The actual start time of each task; For the first Initial preparation time for an unmanned platform; For the first The first unmanned platform to execute the j The processing time required for the first task; for the first task The unmanned platform from the first j The execution location of the first task is transferred to the second task. j +1 processing time required for the execution position of the task; max(.) is the function to find the maximum value; (3.3.2) Calculate the time deviation for each task using the following formula:

[0014] in, For the first j Time deviation of each task; (3.3.3) Calculate the time co-operation factor based on the time deviation of each task. The calculation formula is:

[0015] in, To adjust the parameters.

[0016] Furthermore, the equilibrium factor in step (3.4) The calculation process is as follows: (3.4.1) Calculate the total load of each unmanned platform using the following formula:

[0017] Among them, is the first i Total load of the unmanned platform; For the first i The number of tasks assigned to each unmanned platform; For the first i The unmanned platform was assigned the first... k The processing time required for each task; (3.4.2) Based on the total load of each unmanned platform, calculate the average load using the following formula:

[0018] (3.4.3) Based on the total load and average load of each unmanned platform, calculate the load variance using the following formula:

[0019] (3.4.4) Calculate the balancing factor based on the load variance. The formula is:

[0020] in, This is the proportionality coefficient.

[0021] Furthermore, in step (4), the optimal particle position vector is obtained through the genetic particle swarm optimization algorithm. The specific steps are as follows: (4.1) Calculate the initial population using the multi-objective fitness function. Z t The overall fitness of all particle position vectors is used to determine the optimal particle vector. P t ; (4.2) In the initial population Z t Multiple particle position vectors are randomly selected as parent particle vectors; based on the current optimal particle vector... P t A crossover operation is performed on each parent particle vector to generate the corresponding initial child particle vector; the expression for the crossover operation on the parent particle vectors is:

[0022] in, For the selected number i One parent particle vector; For the first i Each parent particle vector corresponds to a generated initial child particle vector; Indicates an integer crossover operation; (4.3) All initial sub-particle vectors are mutated with a fixed probability, and the resulting sub-particle vectors are obtained after the mutation is completed; (4.4) Calculate the overall fitness of all sub-particle vectors using the multi-objective fitness function; take the sub-particle vector with the highest overall fitness as the optimal sub-particle vector. (4.5) Compare the optimal sub-particle vector with the current optimal particle vector. P t Overall fit; if the overall fit of the optimal sub-particle vector is greater than that of the current optimal particle vector. P t If the overall fit is good, then the best sub-particle vector will be used as a candidate best particle vector; otherwise, the current best particle vector will be used. P t As a candidate optimal particle vector; (4.6) Combine the generated sub-particle vectors with the initial population Z t The particle position vectors of the parent particles from the non-Chinese generation are combined to obtain the new generation population. Z t+1 ; (4.7) Repeat steps (4.1) to (4.6) for a preset number of iterations; each time step (4.1) is executed, the new generation population obtained from the previous iteration is used. Z t+1 As the initial population for this iteration Z t Furthermore, each time the parent particle vector operation in step (4.2) is executed, the expression for the crossover operation is replaced with:

[0023] Among them, is the candidate optimal particle vector with the highest overall fitness among all candidate optimal particle vectors generated in each iteration. (4.8) Among all the candidate optimal particle vectors generated during the iterative process, the candidate optimal particle vector with the highest overall fitness is taken as the optimal particle position vector. .

[0024] Secondly, the present invention also proposes a computer program product, which includes a computer program that, when executed by a processor, implements the steps of the above-mentioned task allocation and time coordination method for a multi-unmanned system.

[0025] Thirdly, the present invention also proposes a processor for running a program, wherein the program executes the above-described method for task allocation and time coordination in a multi-unmanned system.

[0026] Fourthly, the present invention also proposes a storage medium comprising a stored program, wherein, when the program is running, the device containing the storage medium is controlled to execute the above-described method for task allocation and time coordination of a multi-unmanned system.

[0027] The advantages of this invention compared to the prior art are: (1) This invention constructs a multi-objective fitness function, and uses the genetic particle swarm algorithm to continuously iterate and optimize the task allocation scheme of multiple unmanned platforms. In the optimization process, it fully considers various factors such as the performance parameters of multiple unmanned platforms and the comprehensive efficiency during task execution, and finally realizes the task allocation of multiple types of unmanned platforms.

[0028] (2) When constructing the multi-objective fitness function, the present invention includes the capability matching degree between the unmanned platform and the task. The capability matching degree between the unmanned platform and the task is calculated based on the performance parameters of different unmanned platforms, so that the performance indicators of different unmanned platforms can be fully considered when allocating tasks.

[0029] (3) The multi-objective fitness function constructed in this invention includes a time coordination factor, which enables the subsequent algorithm to fully consider the time deviation of multi-platform collaborative tasks during the iterative optimization process, thus significantly improving the overall efficiency.

[0030] (4) This invention continuously iterates and optimizes the task allocation scheme of the unmanned platform through the genetic particle swarm algorithm. By crossover, mutation and retention of the optimal allocation scheme, the algorithm avoids getting trapped in local optima, thereby ensuring that the obtained task allocation scheme is optimal.

[0031] (5) In the iterative optimization process of the genetic particle swarm algorithm, the present invention uses the multi-objective fitness function as the optimization basis, so that the final task allocation result can fully guarantee the balance between task benefits, execution costs, time coordination and load balancing, so that when multiple unmanned platforms perform tasks according to the final scheme, the effect is optimal. Attached Figure Description

[0032] Figure 1 This is a flowchart of a task allocation and time coordination method for a multi-unmanned system according to the present invention. Detailed Implementation

[0033] The specific embodiments of the present invention will now be described in further detail with reference to the accompanying drawings.

[0034] like Figure 1 As shown, this invention provides a method for task allocation and time coordination in a multi-unmanned system, with the following steps: (1) Collection M Status information of heterogeneous unmanned platforms; based on the execution requirements.N Based on the functional requirements of each task and the status information of the unmanned platform, calculate the capability matching degree between each unmanned platform and each task; The status information of the unmanned platform includes payload capacity, maximum speed, detection range, and communication radius.

[0035] The process of calculating the capability matching degree between the unmanned platform and the task in step (1) is as follows: (1.1) Based on the state information of the unmanned platform, extract the state information of each unmanned platform. L Several performance parameters; these performance parameters include payload capacity, maximum speed, detection range, location, speed, and communication radius, etc. (1.2) Each unmanned platform L The performance parameters are arranged in a preset performance parameter order to form the capability vector of the corresponding unmanned platform; among them, the first... i The capability vector of each unmanned platform is , For the first i The first unmanned platform k One performance parameter; (1.3) Based on the functional requirements of each task, statistically analyze the impact of each task on the UAV. L The requirements for each performance parameter are analyzed to construct a requirement vector for each task; where the first parameter is... j The demand vector for each task is: , For the first j The first task for the unmanned platform k Minimum required threshold for each performance parameter; (1.4) Based on the capability vector of each unmanned platform and the requirement vector of the task, calculate the capability matching degree between each unmanned platform and each task. The calculation formula is as follows:

[0036] in, For the first i The unmanned platform and the first j The degree of capability matching between tasks For the first k The weights of each performance parameter are given, and min(.) is the function to find the minimum value.

[0037] (2) Define the particle position vector ,in All are arbitrary integers between 1 and M; the particle position vector X Each element in the table corresponds to a task allocation strategy. Indicates the first i The task is assigned to the first An unmanned platform.

[0038] (3) Based on the capability matching degree between the unmanned platform and the task, a multi-objective fitness function corresponding to the particle position vector is constructed; the calculation result of the multi-objective fitness function represents the overall fitness degree when all tasks are allocated according to the allocation strategy represented by the particle position vector. The specific process of constructing the multi-objective fitness function corresponding to the particle position vector in step (3) is as follows: (3.1) Based on the capability matching degree between the unmanned platform and the task, calculate the task benefit term when all tasks are allocated according to the particle position vector. The calculation formula is:

[0039] in, Indicates the first j Whether the first task has been assigned, when the first task is... j One task was assigned. If it is 1, when the first j One task has not been assigned. =0; For the first j The first task and the first The capability matching degree between task platforms; When constructing the multi-objective fitness function, this invention includes the capability matching degree between unmanned platforms and tasks. The capability matching degree between unmanned platforms and tasks is calculated based on the performance parameters of different unmanned platforms, so that the performance indicators of different unmanned platforms can be fully considered when allocating tasks. (3.2) Calculate the execution cost term when all tasks are assigned according to the particle position vector. The calculation formula is:

[0040] in, and All are weighting coefficients; For the first i The number of tasks assigned to each unmanned platform; For the first i The unmanned platform from the first j The task execution location is transposed to the first... j +Time required to execute 1 task position; For the first i The first unmanned platform to execute the j The processing time required for each task; (3.3) Calculate the time co-operation factor when all tasks are assigned according to the particle position vector. The specific steps are as follows: (3.3.1) Calculate the actual start time of each task after it has been assigned to the corresponding unmanned platform according to the particle position vector. The calculation formula is as follows:

[0041] in, For the first j The actual start time of each task; For the first Initial preparation time for an unmanned platform; For the first The first unmanned platform to execute the j The processing time required for the first task; for the first task The unmanned platform from the first j The execution location of the first task is transferred to the second task. j +1 processing time required for the execution position of the task; max(.) is the function to find the maximum value; (3.3.2) Calculate the time deviation for each task using the following formula:

[0042] in, For the first j Time deviation of each task; (3.3.3) Calculate the time co-operation factor based on the time deviation of each task. The calculation formula is:

[0043] in, To adjust the parameters.

[0044] Through step (3.3), the multi-objective fitness function constructed in this invention includes a time collaboration factor, which enables the subsequent algorithm to fully consider the time deviation of multi-platform collaborative tasks during the iterative optimization process, thereby significantly improving the overall efficiency.

[0045] (3.4) Calculate the balance factor when all tasks are assigned according to the particle position vector. The specific steps are as follows: (3.4.1) Calculate the total load of each unmanned platform using the following formula:

[0046] Among them, is the first i Total load of the unmanned platform; For the first i The number of tasks assigned to each unmanned platform; For the first i The unmanned platform was assigned the first... kThe processing time required for each task; (3.4.2) Based on the total load of each unmanned platform, calculate the average load using the following formula:

[0047] (3.4.3) Based on the total load and average load of each unmanned platform, calculate the load variance using the following formula:

[0048] (3.4.4) Calculate the balancing factor based on the load variance. The formula is:

[0049] in, This is the proportionality coefficient.

[0050] (3.5) Based on task benefit items Execution cost item Time co-factor and balance factor Calculate the particle position vector X Corresponding multi-objective fitness function F The formula is:

[0051] in, , , and These are the task reward items. Execution cost item Time co-factor and balance factor The corresponding weights; For task reward items The theoretical maximum value; For execution cost item The theoretical maximum value.

[0052] (4) Randomly generate multiple particle position vectors, and all particle position vectors constitute the initial population. Z t The initial population is determined using a genetic particle swarm optimization algorithm based on a multi-objective fitness function. Z t Through iteration, the optimal particle position vector is obtained. ; In step (4), the optimal particle position vector is obtained through the genetic particle swarm optimization algorithm. The specific steps are as follows: (4.1) Calculate the initial population using the multi-objective fitness function.Z t The overall fitness of all particle position vectors is used to determine the optimal particle vector. P t ; (4.2) In the initial population Z t Multiple particle position vectors are randomly selected as parent particle vectors; based on the current optimal particle vector... P t A crossover operation is performed on each parent particle vector to generate the corresponding initial child particle vector; the expression for the crossover operation on the parent particle vectors is:

[0053] in, For the selected number i One parent particle vector; For the first i Each parent particle vector corresponds to a generated initial child particle vector; Indicates an integer crossover operation; (4.3) All initial sub-particle vectors are mutated with a fixed probability, and the resulting sub-particle vectors are obtained after the mutation is completed; (4.4) Calculate the overall fitness of all sub-particle vectors using the multi-objective fitness function; take the sub-particle vector with the highest overall fitness as the optimal sub-particle vector. (4.5) Compare the optimal sub-particle vector with the current optimal particle vector. P t Overall fit; if the overall fit of the optimal sub-particle vector is greater than that of the current optimal particle vector. P t If the overall fit is good, then the best sub-particle vector will be used as a candidate best particle vector; otherwise, the current best particle vector will be used. P t As a candidate optimal particle vector; (4.6) Combine the generated sub-particle vectors with the initial population Z t The particle position vectors of the parent particles from the non-Chinese generation are combined to obtain the new generation population. Z t+1 ; (4.7) Repeat steps (4.1) to (4.6) for a preset number of iterations; each time step (4.1) is executed, the new generation population obtained from the previous iteration is used. Z t+1 As the initial population for this iteration Z tFurthermore, each time the parent particle vector operation in step (4.2) is executed, the expression for the crossover operation is replaced with:

[0054] Among them, is the candidate optimal particle vector with the highest overall fitness among all candidate optimal particle vectors generated in each iteration. (4.8) Among all the candidate optimal particle vectors generated during the iterative process, the candidate optimal particle vector with the highest overall fitness is taken as the optimal particle position vector. .

[0055] Based on step (4), this invention continuously iterates and optimizes the task allocation scheme of the unmanned platform through the genetic particle swarm algorithm. By using crossover, mutation and retention of the optimal allocation scheme, the algorithm avoids getting trapped in local optima, thereby ensuring that the obtained task allocation scheme is optimal.

[0056] (5) Assign tasks to the corresponding unmanned platforms according to the allocation strategy corresponding to each element in the optimal particle position vector, and complete the task allocation.

[0057] Based on the above process, this invention constructs a multi-objective fitness function and uses the genetic particle swarm optimization algorithm to iteratively optimize the task allocation scheme for multiple unmanned platforms. During the optimization process, various factors such as the performance parameters of the multiple unmanned platforms and the overall efficiency of task execution are fully considered, ultimately achieving task allocation for multiple types of unmanned platforms. Simultaneously, during the iterative optimization process of the genetic particle swarm optimization algorithm, the multi-objective fitness function is used as the optimization basis, ensuring that the final task allocation result fully guarantees a balance between task benefits, execution costs, time coordination, and load balancing. This allows the multiple unmanned platforms to achieve optimal performance when executing tasks according to the final scheme.

[0058] Secondly, the present invention also proposes a computer program product, which includes a computer program that, when executed by a processor, implements the steps of the above-mentioned task allocation and time coordination method for a multi-unmanned system.

[0059] Thirdly, the present invention also proposes a processor for running a program, wherein the program executes the above-described method for task allocation and time coordination in a multi-unmanned system.

[0060] Fourthly, the present invention also proposes a storage medium comprising a stored program, wherein, when the program is running, the device containing the storage medium is controlled to execute the above-described method for task allocation and time coordination of a multi-unmanned system.

[0061] The parts of this invention not described in detail are common knowledge to those skilled in the art.

Claims

1. A method for task allocation and time coordination in a multi-unmanned system, characterized in that... The steps are as follows: (1) Collection M Status information of heterogeneous unmanned platforms; based on the execution requirements. N Based on the functional requirements of each task and the status information of the unmanned platform, calculate the capability matching degree between each unmanned platform and each task; (2) Define the particle position vector ,in All are arbitrary integers between 1 and M; the particle position vector X Each element in the table corresponds to a task allocation strategy. Indicates the first i The task is assigned to the first An unmanned platform; (3) Based on the capability matching degree between the unmanned platform and the task, a multi-objective fitness function corresponding to the particle position vector is constructed; the calculation result of the multi-objective fitness function represents the overall fitness degree when all tasks are allocated according to the allocation strategy represented by the particle position vector. (4) Randomly generate multiple particle position vectors, and all particle position vectors constitute the initial population. Z t The initial population is determined using a genetic particle swarm optimization algorithm based on a multi-objective fitness function. Z t Through iteration, the optimal particle position vector is obtained. ; (5) Assign tasks to the corresponding unmanned platforms according to the allocation strategy corresponding to each element in the optimal particle position vector, and complete the task allocation.

2. The method for task allocation and time coordination in a multi-unmanned system according to claim 1, characterized in that: The status information of the unmanned platform includes payload capacity, maximum speed, detection range, and communication radius.

3. The method for task allocation and time coordination in a multi-unmanned system according to claim 1, characterized in that: The process of calculating the capability matching degree between the unmanned platform and the task in step (1) is as follows: (1.1) Based on the state information of the unmanned platform, extract the state information of each unmanned platform. L One performance parameter; (1.2) Each unmanned platform L The performance parameters are arranged in a preset performance parameter order to form the capability vector of the corresponding unmanned platform; among them, the first... i The capability vector of each unmanned platform is , For the first i The first unmanned platform k One performance parameter; (1.3) Based on the functional requirements of each task, statistically analyze the impact of each task on the UAV. L The requirements for each performance parameter are analyzed to construct a requirement vector for each task; where the first parameter is... j The demand vector for each task is: , For the first j The first task for the unmanned platform k Minimum required threshold for each performance parameter; (1.4) Based on the capability vector of each unmanned platform and the requirement vector of the task, calculate the capability matching degree between each unmanned platform and each task. The calculation formula is as follows: in, For the first i The unmanned platform and the first j The degree of capability matching between tasks For the first k The weights of each performance parameter are given, and min(.) is the function to find the minimum value.

4. The method for task allocation and time coordination in a multi-unmanned system according to claim 1, characterized in that: The specific process of constructing the multi-objective fitness function corresponding to the particle position vector in step (3) is as follows: (3.1) Based on the capability matching degree between the unmanned platform and the task, calculate the task benefit term when all tasks are allocated according to the particle position vector. The calculation formula is: in, Indicates the first j Whether the first task has been assigned, when the first task is... j One task was assigned. If it is 1, when the first j One task has not been assigned. =0; For the first j The first task and the first The compatibility of capabilities between different task platforms; (3.2) Calculate the execution cost term when all tasks are assigned according to the particle position vector. The calculation formula is: in, and All are weighting coefficients; For the first i The number of tasks assigned to each unmanned platform; For the first i The unmanned platform from the first j The task execution location is transposed to the first... j +Time required to execute 1 task position; For the first i The first unmanned platform to execute the j The processing time required for each task; (3.3) Calculate the time co-operation factor when all tasks are assigned according to the particle position vector. ; (3.4) Calculate the balance factor when all tasks are assigned according to the particle position vector. ; (3.5) Based on task benefit items Execution cost item Time co-factor and balance factor Calculate the particle position vector X Corresponding multi-objective fitness function F The formula is: in, , , and These are the task reward items. Execution cost item Time co-factor and balance factor The corresponding weights; For task reward items The theoretical maximum value; For execution cost item The theoretical maximum value.

5. The method for task allocation and time coordination in a multi-unmanned system according to claim 4, characterized in that: The time coordination factor in step (3.3) The calculation process is as follows: (3.3.1) Calculate the actual start time of each task after it has been assigned to the corresponding unmanned platform according to the particle position vector. The calculation formula is as follows: in, For the first j The actual start time of each task; For the first Initial preparation time for an unmanned platform; For the first The first unmanned platform to execute the j The processing time required for the first task; for the first task The unmanned platform from the first j The execution location of the first task is transferred to the second task. j +1 processing time required for the execution position of the task; max(.) is the function to find the maximum value; (3.3.2) Calculate the time deviation for each task using the following formula: in, For the first j Time deviation of each task; (3.3.3) Calculate the time co-operation factor based on the time deviation of each task. The calculation formula is: in, To adjust the parameters.

6. The method for task allocation and time coordination in a multi-unmanned system according to claim 4, characterized in that: The balancing factor in step (3.4) The calculation process is as follows: (3.4.1) Calculate the total load of each unmanned platform using the following formula: Among them, is the first i Total load of the unmanned platform; For the first i The number of tasks assigned to each unmanned platform; For the first i The unmanned platform was assigned the first... k The processing time required for each task; (3.4.2) Based on the total load of each unmanned platform, calculate the average load using the following formula: (3.4.3) Based on the total load and average load of each unmanned platform, calculate the load variance using the following formula: (3.4.4) Calculate the balancing factor based on the load variance. The formula is: in, This is the proportionality coefficient.

7. The method for task allocation and time coordination in a multi-unmanned system according to claim 1, characterized in that: In step (4), the optimal particle position vector is obtained through the genetic particle swarm optimization algorithm. The specific steps are as follows: (4.1) Calculate the initial population using the multi-objective fitness function. Z t The overall fitness of all particle position vectors is used to determine the optimal particle vector. P t ; (4.2) In the initial population Z t Multiple particle position vectors are randomly selected as parent particle vectors; based on the current optimal particle vector... P t A crossover operation is performed on each parent particle vector to generate the corresponding initial child particle vector; the expression for the crossover operation on the parent particle vectors is: in, For the selected number i One parent particle vector; For the first i Each parent particle vector corresponds to a generated initial child particle vector; Indicates an integer crossover operation; (4.3) All initial sub-particle vectors are mutated with a fixed probability, and the resulting sub-particle vectors are obtained after the mutation is completed; (4.4) Calculate the overall fitness of all sub-particle vectors using the multi-objective fitness function; take the sub-particle vector with the highest overall fitness as the optimal sub-particle vector. (4.5) Compare the optimal sub-particle vector with the current optimal particle vector. P t Overall fit; if the overall fit of the optimal sub-particle vector is greater than that of the current optimal particle vector. P t If the overall fit is good, then the best sub-particle vector will be used as a candidate best particle vector; otherwise, the current best particle vector will be used. P t As a candidate optimal particle vector; (4.6) Combine the generated sub-particle vectors with the initial population Z t The particle position vectors of the parent particles from the non-Chinese generation are combined to obtain the new generation population. Z t+1 ; (4.7) Repeat steps (4.1) to (4.6) for a preset number of iterations; each time step (4.1) is executed, the new generation population obtained from the previous iteration is used. Z t+1 As the initial population for this iteration Z t Furthermore, each time the parent particle vector operation in step (4.2) is executed, the expression for the crossover operation is replaced with: Among them, is the candidate optimal particle vector with the highest overall fitness among all candidate optimal particle vectors generated in each iteration. (4.8) Among all the candidate optimal particle vectors generated during the iterative process, the candidate optimal particle vector with the highest overall fitness is taken as the optimal particle position vector. .

8. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the steps of the task allocation and time coordination method for a multi-unmanned system as described in any one of claims 1 to 7.

9. A processor, characterized in that, The processor is used to run a program, wherein the program executes a task allocation and time coordination method for a multi-unmanned system according to any one of claims 1 to 7.

10. A storage medium, characterized in that, The storage medium includes a stored program, wherein, when the program is running, it controls the device where the storage medium is located to execute the task allocation and time coordination method of any one of claims 1 to 7.

Images