CN-121979232-A - Task allocation and time coordination method for multi-unmanned system

CN121979232ACN 121979232 ACN121979232 ACN 121979232ACN-121979232-A

Abstract

A task allocation and time coordination method for a multi-unmanned system is characterized by comprising the following steps of (1) calculating capability matching degree based on function requirement conditions of tasks to be executed and combining state information of unmanned platforms, (2) defining particle position vectors to represent task allocation strategies, (3) constructing multi-target fitness functions corresponding to the particle position vectors to represent overall fitness of all tasks when the tasks are allocated according to the particle position vectors based on the capability matching degree between the unmanned platforms and the tasks, (4) constructing an initial population, carrying out iteration of the initial population according to the multi-target fitness functions through a genetic particle swarm algorithm to obtain optimal particle position vectors, and (5) allocating the tasks to the corresponding unmanned platforms according to the optimal particle position vectors to complete task allocation. Through the process, the invention fully considers various factors such as the performance parameters of the unmanned platform, the comprehensive efficiency during task execution and the like in the process of optimizing the task allocation scheme.

Inventors

HU CHANGQING
NIU GENYUAN
XIE JIAWEN
CAI LIMING
WANG XIAO
DING SHENGNAN
WANG XINYUE
XU RUNYU

Assignees

北京航天控制仪器研究所

Dates

Publication Date: 20260505
Application Date: 20251229

Claims (10)

1. A task allocation and time coordination method of a multi-unmanned system is characterized by comprising the following steps: (1) Based on the function requirement conditions of N tasks to be executed, combining the state information of the unmanned platforms, and calculating the capability matching degree between each unmanned platform and each task; (2) Defining particle position vectors Wherein Each element in the particle position vector X corresponds to the allocation strategy of the task, Indicating assignment of the ith task to the ith A personal unmanned platform; (3) Constructing a multi-objective fitness function corresponding to the particle position vector based on the capability matching degree between the unmanned platform and the task, wherein the calculation result of the multi-objective fitness function represents the overall fitness of all the tasks when the tasks are distributed according to the distribution strategy represented by the particle position vector; (4) Randomly generating a plurality of particle position vectors, wherein all the particle position vectors form an initial population Z t , iterating the initial population Z t by using a genetic particle swarm algorithm based on a multi-target fitness function to obtain an optimal particle position vector ; (5) And distributing the tasks to the corresponding unmanned platforms according to the distribution strategy corresponding to each element in the optimal particle position vector, and completing task distribution.
2. The method of claim 1, wherein the status information of the unmanned platform comprises load capacity, maximum speed, detection distance and communication radius.
3. The method for task allocation and time coordination of the multi-unmanned system according to claim 1, wherein the capability matching degree calculation process between the unmanned platform and the task in the step (1) is as follows: Based on the state information of the unmanned platforms, extracting L performance parameters of each unmanned platform; (1.2) arranging the L performance parameters of each unmanned platform according to a preset performance parameter sequence to form a capacity vector of the corresponding unmanned platform, wherein the capacity vector of the ith unmanned platform is , The kth performance parameter of the ith unmanned platform; (1.3) based on the function requirement situation of each task, counting the requirement situation of each task on the performance parameters of the unmanned plane L, thereby constructing a requirement vector of each task, wherein the requirement vector of the j-th task is , A minimum requirement threshold for the kth performance parameter of the unmanned platform for the jth task; (1.4) calculating the capability matching degree between each unmanned platform and each task based on the capability vector of each unmanned platform and the requirement vector of the task, wherein the calculation formula is as follows: Wherein, the For capability matching between the ith unmanned platform and the jth task, For the weight of the kth performance parameter, min () is a minimum function.
4. The method for task allocation and time coordination of a multi-unmanned system according to claim 1, wherein the specific process of constructing the multi-objective fitness function corresponding to the particle position vector in the step (3) is as follows: (3.1) calculating task benefit items when all tasks are distributed according to the particle position vector based on the capability matching degree between the unmanned platform and the tasks The calculation formula is Wherein, the Indicating whether or not the jth task is assigned, and when the jth task is assigned, 1, When the j-th task is not assigned, Is 0; for the jth task and the jth Capability matching degree among the task platforms; (3.2) calculating the execution cost term when all tasks are allocated according to the particle position vector The calculation formula is Wherein, the And Are all weight coefficients; The number of tasks assigned to the ith unmanned platform; the time required for the ith unmanned platform to transpose from the jth task execution position to the jth+1th task execution position; processing time required to perform the jth task for the ith unmanned platform; (3.3) calculating a time cofactor for all tasks when they are distributed according to the particle position vector ; (3.4) Calculating an equalization factor when all tasks are distributed according to the particle position vector ; (3.5) Task-based revenue items Execution cost term Time co-factor Equalization factor And calculating a multi-target fitness function F corresponding to the particle position vector X, wherein the formula is as follows: Wherein, the 、、 And Respectively, task income items Execution cost term Time co-factor Equalization factor Corresponding weights; for task benefit Is a theoretical maximum of (2); to execute the cost item Is not shown in the drawing).
5. The method for task allocation and time coordination of multiple unmanned systems according to claim 4, wherein the time coordination factor in step (3.3) is The calculation process of (1) is as follows: (3.3.1) calculating the actual starting time of each task after each task is allocated to the corresponding unmanned platform according to the particle position vector, wherein the calculation formula is as follows: Wherein, the The actual start time for the jth task; Is the first Initial preparation completion time of the personal unmanned platform; Is the first The processing time required by the personal unmanned platform for executing the j-th task is the j-th task The processing time required by the personal unmanned platform to transfer from the execution position of the jth task to the execution position of the (j+1) th task is max (number) which is a maximum function; (3.3.2) calculating the time deviation of each task, wherein the calculation formula is as follows: Wherein, the Time deviation for the j-th task; (3.3.3) calculating a time cofactor based on the time bias of each task The calculation formula is as follows: Wherein, the To adjust the parameters.
6. The method of task allocation and time coordination of multiple unmanned systems according to claim 4, wherein the equalization factor in step (3.4) The calculation process of (1) is as follows: (3.4.1) calculating the total load of each unmanned platform, wherein the calculation formula is as follows: wherein, is the total load of the ith unmanned platform; The number of tasks assigned to the ith unmanned platform; Processing time required to execute the assigned kth task for the ith unmanned platform; (3.4.2) calculating an average load based on the total load of each unmanned platform, the formula being: (3.4.3) calculating a load variance based on the total load and the average load of each unmanned platform, the formula being: (3.4.4) calculating an equalization factor based on the load variance The formula is: Wherein, the Is a proportionality coefficient.
7. The method of task allocation and time coordination of unmanned aerial vehicle system according to claim 1, wherein in step (4), the optimal particle position vector is obtained by genetic particle swarm algorithm The specific steps of (a) are as follows: (4.1) calculating the overall fitness corresponding to all the particle position vectors in the initial population Z t by using a multi-target fitness function, wherein the particle position vector with the highest overall fitness is used as a current optimal particle vector P t ; (4.2) randomly selecting a plurality of particle position vectors as parent particle vectors in the initial population Z t , performing cross operation on each parent particle vector based on the current optimal particle vector P t to generate a corresponding initial child particle vector, wherein the expression of performing cross operation on the parent particle vectors is as follows: Wherein, the For the selected ith parent particle vector; An initial sub-particle vector correspondingly generated for the ith parent particle vector; representing an integer crossover operation; (4.3) mutating all the initial particle vectors with fixed probability, and obtaining the particle vectors after the mutation is finished; (4.4) calculating the overall fitness of all the sub-particle vectors by utilizing a multi-target fitness function, wherein the sub-particle vector with the highest overall fitness is used as the optimal sub-particle vector; (4.5) comparing the overall fitness of the optimal sub-particle vector with the current optimal particle vector P t , wherein if the overall fitness of the optimal sub-particle vector is greater than that of the current optimal particle vector P t , the optimal sub-particle vector is used as a candidate optimal particle vector, otherwise, the current optimal particle vector P t is used as a candidate optimal particle vector; (4.6) combining the generated sub-particle vector with the particle position vector of the non-parent particle vector in the initial population Z t to obtain a new generation population Z t+1 ; Steps (4.1) - (4.6) of the preset iteration times are circularly executed, a new generation population Z t+1 obtained by the previous iteration is used as an initial population Z t of the iteration every time the step (4.1) is executed, and the expression of the cross operation is replaced by the expression of the cross operation every time the parent particle vector operation in the step (4.2) is executed: among all candidate optimal particle vectors generated in the previous iteration process, the candidate optimal particle vector with the highest overall adaptation degree is selected; (4.8) using the candidate optimal particle vector with the highest overall adaptation degree of the candidate optimal particle vectors generated in all loop iteration processes as the optimal particle position vector 。
8. A computer program product, characterized in that the computer program product comprises a computer program which, when executed by a processor, implements the steps of a method for task allocation and time coordination of a multiple unmanned system according to any of claims 1 to 7.
9. A processor, wherein the processor is configured to run a program, and wherein the program executes a task allocation and time coordination method of the multi-unmanned system according to any of claims 1 to 7.
10. A storage medium, characterized in that the storage medium comprises a stored program, wherein the program, when run, controls a device in which the storage medium is located to execute a task allocation and time coordination method of the multi-unmanned system according to any one of claims 1 to 7.

Description

Task allocation and time coordination method for multi-unmanned system Technical Field The invention relates to a task allocation and time coordination method for a multi-unmanned system, and belongs to the technical field of unmanned system coordination control. Background With the continuous development of unmanned system technology, multiple types of unmanned platforms, such as unmanned aerial vehicles, unmanned boats and unmanned ships, for cooperatively executing tasks have become an important development trend. However, the task allocation of the existing multi-unmanned platform is mainly aimed at a single type of unmanned platform, so that task execution of different types of unmanned platforms is difficult to be effectively and comprehensively allocated according to performance parameters of different unmanned platforms, meanwhile, the existing method only usually considers a single optimization target when task allocation is carried out, and comprehensive consideration of factors such as comprehensive efficiency and balanced load of different unmanned platforms when task execution is difficult to be comprehensively considered. Disclosure of Invention The technical problem of the invention is to overcome the defects of the prior art and provide a task allocation and time coordination method of a multi-unmanned system. According to the method, a multi-objective fitness function is constructed, a genetic particle swarm algorithm is utilized to continuously iterate and optimize a multi-unmanned platform task allocation scheme, various factors such as performance parameters of the multi-unmanned platform and comprehensive efficiency during task execution are fully considered in the optimization process, and the problem that only a single optimization objective is considered when task allocation is carried out on a single type of unmanned platform in the conventional method is solved. The technical scheme of the invention is as follows: A task allocation and time cooperation method of a multi-unmanned system comprises the following steps: (1) Based on the function requirement conditions of N tasks to be executed, combining the state information of the unmanned platforms, and calculating the capability matching degree between each unmanned platform and each task; (2) Defining particle position vectors WhereinEach element in the particle position vector X corresponds to the allocation strategy of the task,Indicating assignment of the ith task to the ithA personal unmanned platform; (3) Constructing a multi-objective fitness function corresponding to the particle position vector based on the capability matching degree between the unmanned platform and the task, wherein the calculation result of the multi-objective fitness function represents the overall fitness of all the tasks when the tasks are distributed according to the distribution strategy represented by the particle position vector; (4) Randomly generating a plurality of particle position vectors, wherein all the particle position vectors form an initial population Z t, iterating the initial population Z t by using a genetic particle swarm algorithm based on a multi-target fitness function to obtain an optimal particle position vector ; (5) And distributing the tasks to the corresponding unmanned platforms according to the distribution strategy corresponding to each element in the optimal particle position vector, and completing task distribution. Further, the status information of the unmanned platform comprises load capacity, maximum navigational speed, detection distance and communication radius. Further, the capability matching degree calculation process between the unmanned platform and the task in the step (1) is as follows: Based on the state information of the unmanned platforms, extracting L performance parameters of each unmanned platform; (1.2) arranging the L performance parameters of each unmanned platform according to a preset performance parameter sequence to form a capacity vector of the corresponding unmanned platform, wherein the capacity vector of the ith unmanned platform is ,The kth performance parameter of the ith unmanned platform; (1.3) based on the function requirement situation of each task, counting the requirement situation of each task on the performance parameters of the unmanned plane L, thereby constructing a requirement vector of each task, wherein the requirement vector of the j-th task is ,A minimum requirement threshold for the kth performance parameter of the unmanned platform for the jth task; (1.4) calculating the capability matching degree between each unmanned platform and each task based on the capability vector of each unmanned platform and the requirement vector of the task, wherein the calculation formula is as follows: Wherein, the For capability matching between the ith unmanned platform and the jth task,For the weight of the kth performance parameter, min () is a minimum function. Further, the specifi