A distributed cooperative scheduling method, device, system, and storage medium for unmanned swarms to perform multi-modal heterogeneous dynamic tasks.

Abstract

This invention discloses a distributed collaborative scheduling method, apparatus, system, and storage medium for unmanned swarms to perform multi-modal heterogeneous dynamic tasks. Belonging to the field of unmanned swarm collaborative scheduling technology, it includes: in response to task changes, allocating multi-modal heterogeneous dynamic tasks according to preset dynamic task collaborative scheduling rules until all unmanned swarm members no longer wish to change their decisions to improve returns; and in response to no task changes, allocating multi-modal heterogeneous dynamic tasks according to preset static task collaborative scheduling rules until all unmanned swarm members no longer wish to change their decisions to improve returns. This invention can improve resource utilization and increase the total returns and average success rate of the unmanned swarm system.

Description

Technical Field

[0001] This invention relates to a distributed collaborative scheduling method, apparatus, system, and storage medium for unmanned swarms to perform multi-modal heterogeneous dynamic tasks, belonging to the field of unmanned swarm collaborative scheduling technology. Background Technology

[0002] With the rapid development of intelligent unmanned swarm technology, the platform's diverse characteristics, good resource allocation space, and functional flexibility have made task allocation in unmanned swarms a research hotspot and frontier, with many scholars making significant contributions. Currently, unmanned swarm task allocation methods can be mainly divided into centralized and distributed algorithms, with cooperative game theory algorithms holding a dominant position in distributed task allocation. Addressing the problems of uneven task allocation, poor coordination, and weak dynamic adaptability in unmanned swarm systems performing multi-objective tasks, a key challenge is how to jointly consider the value, priority, and requirements of multiple coupled heterogeneous tasks, as well as the impact of task changes, to achieve distributed collaborative scheduling of resources from different swarm members (UAVs and unmanned vehicles), realizing a reasonable match between the resources of heterogeneous swarm members and heterogeneous tasks, and enabling efficient dynamic adjustments based on task changes.

[0003] Current methods often assume that task requirements are singular. However, in reality, coupling relationships between tasks are common. For example, in reconnaissance operations, image acquisition and computational processing, as well as information transmission, are coupled and interdependent; the lack of any corresponding resource can lead to poor reconnaissance mission performance. Therefore, finer-grained task and resource allocation is needed to achieve deeper collaborative optimization. Furthermore, current methods largely overlook situations where cluster members can simultaneously allocate resources to multiple heterogeneous tasks, or where a single task can be collaboratively completed by multiple cluster members. As the complexity caused by task heterogeneity increases, and the capabilities and resources of members diversify due to enhanced intelligence, it is necessary to further investigate the problem of complex resource allocation to improve the resource utilization rate of cluster collaboration. Summary of the Invention

[0004] The purpose of this invention is to provide a distributed collaborative scheduling method, device, system, and storage medium for unmanned clusters to perform multi-modal heterogeneous dynamic tasks, thereby solving the problem of low resource utilization in the prior art.

[0005] To achieve the above objectives, the present invention employs the following technical solution:

[0006] In a first aspect, the present invention provides a distributed cooperative scheduling method for unmanned swarms to perform multi-modal heterogeneous dynamic tasks, comprising:

[0007] In response to changes in tasks, multi-mode heterogeneous dynamic tasks are allocated according to preset dynamic task collaborative scheduling rules until all unmanned cluster members no longer want to change their own decisions to improve benefits.

[0008] In response to the unchanged task, multi-mode heterogeneous dynamic tasks are allocated according to the preset static task collaborative scheduling rules until all unmanned cluster members no longer want to change their own decisions to improve their benefits.

[0009] In conjunction with the first aspect, further, before allocating multi-modal heterogeneous dynamic tasks according to dynamic / static task collaborative scheduling rules, the following steps are also included:

[0010] Obtain the initial information of multi-mode heterogeneous dynamic tasks and unmanned cluster members, and randomly allocate the multi-mode heterogeneous dynamic tasks accordingly to obtain the alliance under the initial allocation scheme.

[0011] Calculate the total revenue and average success rate of the alliance structure under the initial allocation scheme. The total revenue and average success rate of the alliance structure are used to provide data judgment basis for dynamic / static task collaborative scheduling rules.

[0012] In conjunction with the first aspect, further, the total benefit of the alliance structure is the sum of the benefits of all alliances, the benefit of an alliance is the sum of the benefits of its members, and the benefit of each member is utility minus cost.

[0013] The alliance's revenue is calculated using the following formula:

[0014]

[0015] Among them, U m (Task) represents the reward for the alliance to execute multi-mode heterogeneous dynamic task m, where n is the drone and g is the unmanned vehicle. The benefits for alliance member drones n performing multi-mode heterogeneous dynamic tasks m. The benefits for autonomous vehicles g performing multi-modal heterogeneous dynamic tasks m for alliance members. For a collection of UAVs in a multi-mode heterogeneous dynamic task m, For the set of unmanned vehicles in the multimodal heterogeneous dynamic task m;

[0016]

[0017]

[0018] Among them, R n R g C n R g The utility and cost of performing multimodal heterogeneous dynamic tasks by alliance member drones n and unmanned vehicles g, respectively;

[0019]

[0020] Among them, U m (S) represents the total revenue of the alliance structure, m represents the multi-modal heterogeneous dynamic task, and M represents the set of multi-modal heterogeneous dynamic tasks.

[0021] In conjunction with the first aspect, the average success rate of the alliance structure is further calculated using the following formula:

[0022]

[0023] Where Suc is the average success rate of the alliance structure, L is the number of the multimodal heterogeneous dynamic tasks, and suc m The success rate of executing the multimodal heterogeneous dynamic task m is given by M, where M is the set of multimodal heterogeneous dynamic tasks.

[0024] In conjunction with the first aspect, further, in response to a change in tasks, the allocation of multi-modal heterogeneous dynamic tasks according to preset dynamic task collaborative scheduling rules includes:

[0025] In response to a change in tasks, retrieve the multi-modal heterogeneous dynamic tasks that occurred in both the current and previous time steps. Members of the alliances of these multi-modal heterogeneous dynamic tasks leave their respective alliances. Then, randomly assign the multi-modal heterogeneous dynamic tasks at the current time step and execute the following loop:

[0026] The unmanned cluster members make decisions to obtain a new alliance structure. In response to the unmanned cluster members meeting the preset alliance change rules, the alliance structure is updated under the rules, and the total revenue and average success rate of the alliance structure are updated.

[0027] In conjunction with the first aspect, further, in response to the unchanged task, the allocation of multi-modal heterogeneous dynamic tasks according to preset static task collaborative scheduling rules includes:

[0028] If the task remains unchanged, execute the following loop:

[0029] The unmanned cluster members make decisions to obtain a new alliance structure. In response to the unmanned cluster members meeting the preset alliance change rules, the alliance structure is updated under the rules, and the total revenue and average success rate of the alliance structure are updated.

[0030] In conjunction with the first aspect, further, the step of responding to unmanned cluster members satisfying a preset alliance change rule and executing an update of the alliance structure under that rule includes:

[0031] If joining a consortium increases the total revenue of the consortium and the revenue gained by the unmanned cluster member, then the unmanned cluster member joins the consortium.

[0032] If the total revenue of the alliance structure and the revenue obtained by the unmanned cluster member do not decrease after the unmanned cluster member leaves the alliance, then the unmanned cluster member leaves the alliance.

[0033] If the total revenue of the alliance structure increases after a member of an unmanned cluster moves from its current alliance to another alliance, and the sum of the revenues of the two alliances involved after the move is greater than the sum of the revenues of the two alliances involved before the move, then the member of the unmanned cluster moves from its current alliance to another alliance.

[0034] Secondly, the present invention also provides a distributed cooperative scheduling device for unmanned swarms to perform multi-modal heterogeneous dynamic tasks, comprising:

[0035] Task dynamic scheduling module: In response to changes in tasks, it allocates multi-mode heterogeneous dynamic tasks according to preset dynamic task collaborative scheduling rules until all unmanned cluster members no longer want to change their own decisions to improve benefits.

[0036] The static task scheduling module is used to allocate multi-mode heterogeneous dynamic tasks according to preset static task collaborative scheduling rules when the task has not changed, until all unmanned cluster members no longer want to change their own decisions to improve their benefits.

[0037] Thirdly, the present invention also provides a distributed cooperative scheduling system for unmanned swarms to perform multi-modal heterogeneous dynamic tasks, comprising:

[0038] Memory, used to store instructions;

[0039] A processor is configured to execute the instructions, causing the device to perform operations implementing a distributed cooperative scheduling method for unmanned swarms to perform multi-modal heterogeneous dynamic tasks as described in any of the first aspects.

[0040] Fourthly, the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the operation of a distributed cooperative scheduling method for unmanned clusters to perform multi-modal heterogeneous dynamic tasks as described in any of the first aspects.

[0041] Compared with the prior art, the beneficial effects achieved by the present invention are:

[0042] This invention provides a distributed collaborative scheduling method, device, system, and storage medium for unmanned swarms to perform multi-modal heterogeneous dynamic tasks. It addresses the fine-grained matching optimization problem between multi-modal heterogeneous dynamic tasks with heterogeneous value, heterogeneous requirements, and coupled relationships and the heterogeneous resources of unmanned swarm members, achieving a deeper level of collaborative optimization. For dynamically changing task situations, it performs efficient dynamic adjustments and convergence based on the changes in task conditions, achieving stability of the alliance structure, improving the total benefits and average success rate of the unmanned swarm system, and increasing resource utilization. Attached Figure Description

[0043] Figure 1 This is one of the flowcharts of a distributed collaborative scheduling method for unmanned clusters to perform multi-modal heterogeneous dynamic tasks provided in an embodiment of the present invention;

[0044] Figure 2 This is the second flowchart of a distributed collaborative scheduling method for unmanned clusters to perform multi-modal heterogeneous dynamic tasks provided in an embodiment of the present invention;

[0045] Figure 3 This is a simulation diagram of the alliance after stabilization using the method of the present invention, provided in an embodiment of the present invention.

[0046] Figure 4 This is a comparison chart of the cumulative revenue of the alliance structure under different algorithms provided in the embodiments of the present invention;

[0047] Figure 5 This is a comparison chart of the total revenue per second of the alliance structures formed under different algorithms provided in the embodiments of the present invention;

[0048] Figure 6 This is a comparison chart of the average success rate of alliance structures under different algorithms provided in the embodiments of the present invention. Detailed Implementation

[0049] The present invention will be further described below with reference to the accompanying drawings. The following embodiments are only used to illustrate the technical solution of the present invention more clearly, and should not be used to limit the scope of protection of the present invention.

[0050] Example 1

[0051] like Figure 1 As shown, this embodiment of the invention provides a distributed cooperative scheduling method for unmanned clusters to perform multi-modal heterogeneous dynamic tasks, including the following steps:

[0052] S1. Obtain the initial information of multi-mode heterogeneous dynamic tasks and unmanned cluster members, and randomly allocate the multi-mode heterogeneous dynamic tasks accordingly to obtain the alliance under the initial allocation scheme.

[0053] S2. Calculate the total revenue and average success rate of the alliance structure under the initial allocation scheme. The total revenue and average success rate of the alliance structure are used to provide data judgment basis for dynamic / static task collaborative scheduling rules.

[0054] The total benefit of a coalition structure is the sum of the benefits of all coalitions. The benefit of a coalition is the sum of the benefits of its members. The benefit of a coalition member is utility minus cost.

[0055] The alliance's revenue is calculated using the following formula:

[0056]

[0057] Among them, U m (Task) represents the reward for the alliance to execute multi-mode heterogeneous dynamic task m, where n is the drone and g is the unmanned vehicle. The benefits for alliance member drones n performing multi-mode heterogeneous dynamic tasks m. The benefits for autonomous vehicles g performing multi-modal heterogeneous dynamic tasks m for alliance members. For a collection of UAVs in a multi-mode heterogeneous dynamic task m, For the set of unmanned vehicles in the multimodal heterogeneous dynamic task m;

[0058]

[0059]

[0060] Among them, R n R g C n C g The utility and cost of performing multimodal heterogeneous dynamic tasks by alliance member drones n and unmanned vehicles g, respectively;

[0061]

[0062] Among them, U m (S) represents the total revenue of the alliance structure, m represents the multi-modal heterogeneous dynamic task, and M represents the set of multi-modal heterogeneous dynamic tasks.

[0063] The average success rate of alliance structures is calculated using the following formula:

[0064]

[0065] Where Suc is the average success rate of the alliance structure, L is the number of multi-modal heterogeneous dynamic tasks, and suc m The success rate of executing a multimodal heterogeneous dynamic task m is given by M, where M is the set of multimodal heterogeneous dynamic tasks.

[0066] S3. In response to changes in tasks, allocate multi-mode heterogeneous dynamic tasks according to preset dynamic task collaborative scheduling rules until all unmanned cluster members no longer want to change their own decisions to improve benefits.

[0067] When a task changes, retrieve the tasks that appeared at this moment and the previous moment. The alliance members of these tasks leave their respective alliances. The task at this moment is randomly assigned. The cycle continues: drones and unmanned vehicles (unmanned swarm members) make decisions to obtain a new alliance structure. If the alliance joining, leaving, and transfer rules are met, update the alliance structure, the total benefit of the task coordination alliance structure, and the success rate. The cycle ends when no drone or unmanned vehicle wants to unilaterally change its decision to increase the benefit.

[0068] S4. In response to the fact that the task has not changed, the multi-mode heterogeneous dynamic task is allocated according to the preset static task collaborative scheduling rules until all unmanned cluster members no longer want to change their own decisions to improve the benefits.

[0069] When the task remains unchanged, the loop continues: drones and unmanned vehicles (unmanned swarm members) make decisions to obtain a new alliance structure. If the rules for joining, leaving, and transferring the alliance are met, the alliance structure, the total benefit of the task coordination alliance structure, and the success rate are updated. The loop ends when no drone or unmanned vehicle wants to unilaterally change its decision to increase the benefit.

[0070] The alliance's rule changes include:

[0071] If joining a consortium increases the total revenue of the consortium and the revenue gained by the unmanned cluster member, then the unmanned cluster member joins the consortium.

[0072] If the total revenue of the alliance structure and the revenue obtained by the unmanned cluster member do not decrease after the unmanned cluster member leaves the alliance, then the unmanned cluster member leaves the alliance.

[0073] If the total revenue of the alliance structure increases after a member of an unmanned cluster moves from its current alliance to another alliance, and the sum of the revenues of the two alliances involved after the move is greater than the sum of the revenues of the two alliances involved before the move, then the member of the unmanned cluster moves from its current alliance to another alliance.

[0074] Example 2

[0075] like Figure 2 As shown in the figure, this embodiment of the invention also provides a distributed cooperative scheduling method for unmanned clusters to perform multi-modal heterogeneous dynamic tasks, including the following steps:

[0076] Step 1: Obtain information on heterogeneous tasks and unmanned cluster members (drones and unmanned vehicles) at the initial moment, and randomly assign them to form an initial allocation scheme.

[0077] Step 2: Calculate the total revenue and average success rate of the alliance at the initial moment.

[0078] In this embodiment, the application scenario is set up so that at the beginning, there are 3 tasks, 15 drones, and 3 unmanned vehicles randomly distributed in an area of ​​1000m*1000m. As time goes on, new task objectives will appear.

[0079] The set of drones is defined as N = {1, 2, ..., n}, the set of unmanned vehicles is defined as G = {1, 2, ..., g}, and the set of tasks is defined as M = {1, 2, ..., m}. Assume that the set of tasks appearing at the initial time step is M. I , Drones and unmanned vehicles allocate some resources to the tasks that arise at the initial moment. If a new task objective appears in the next moment, the drones and unmanned vehicles will reallocate their resources according to the task situation, and ultimately work together to complete all tasks.

[0080] The total revenue of the alliance structure can be calculated using the following method:

[0081]

[0082] Among them, U m (S) represents the total revenue of the alliance structure, m represents the multimodal heterogeneous dynamic task, M represents the set of multimodal heterogeneous dynamic tasks, and U m (Task) is the reward for the alliance to execute multi-modal heterogeneous dynamic task m.

[0083] The rewards for the mission alliance m are calculated using the following method:

[0084]

[0085] Among them, U m (Task) represents the reward for the alliance to execute multi-mode heterogeneous dynamic task m, where n is the drone and g is the unmanned vehicle. The benefits for alliance member drones n performing multi-mode heterogeneous dynamic tasks m. The benefits for autonomous vehicles g performing multi-modal heterogeneous dynamic tasks m for alliance members. For a collection of UAVs in a multi-mode heterogeneous dynamic task m, For the set of unmanned vehicles in the multimodal heterogeneous dynamic task m;

[0086]

[0087]

[0088] Among them, R n R g C n C g The utility and cost of the alliance member drones n and unmanned vehicles g performing multimodal heterogeneous dynamic tasks are respectively.

[0089] The utility of alliance members can be obtained through the following methods:

[0090]

[0091]

[0092] Among them, V m For the value of the mission alliance m, F m Let ω1 be the priority of the mission alliance m, and D be the balance coefficient. n Assign resources to the mission alliance m for member drones, D g Allocate resources from mission alliance m to the autonomous vehicles of alliance members, S n and S g To improve the success rate of missions performed by member drones and unmanned vehicles.

[0093] The success rate of alliance member drones and unmanned vehicles performing alliance missions can be obtained using the following methods:

[0094]

[0095]

[0096] in, These represent the communication resource requirements, reconnaissance resource requirements, and jamming resource requirements for task m, respectively. n Assign resources to the mission alliance m for member drones, D g Allocate resources to the task alliance m for the autonomous vehicles of alliance members.

[0097] The cost to alliance members can be obtained through the following methods:

[0098]

[0099]

[0100] Where β, λ1, λ2, and λ3 are balance coefficients. The communication power consumption, reconnaissance power consumption, and jamming power consumption of the alliance member UAV n performing mission m. The communication power consumption, reconnaissance power consumption, and jamming power consumption of the unmanned vehicle g performing mission m for alliance members.

[0101] The average success rate of alliance structures can be obtained using the following methods:

[0102]

[0103] Where L is the number of mission alliances, suc m The success rate of the mission alliance m.

[0104] The success rate of mission alliance m can be obtained by the following method:

[0105]

[0106] Among them, suc m For the success rate of the mission alliance m. This represents the communication resources contributed by drone n, a member of the alliance, to mission m. This represents the reconnaissance resources contributed by member drone n to mission m. This represents the interference resources contributed by drone n, a member of the alliance, to mission m. This indicates the communication resources contributed by the autonomous vehicle g from the alliance member to task m. This indicates the reconnaissance resources contributed by the alliance member's unmanned vehicle g to mission m. This indicates the interference resources contributed by the autonomous vehicle g from the alliance member to task m. This indicates the communication relay resources contributed by the autonomous vehicle g from the alliance member to task m. This represents the reconnaissance computing resources contributed by the unmanned vehicle g from the alliance member to mission m. These are the communication resource requirements, reconnaissance resource requirements, and jamming resource requirements for mission coalition m, respectively.

[0107] Step 3: Drones and unmanned vehicles make decisions, allocating their carried resources to tasks that occur at any given moment, until all tasks are assigned. If a task remains unchanged at a given moment, the drone swarm members allocate resources according to the alliance rules until none of the members meet the alliance rules at the time of decision-making, then the next task is assigned until all tasks are assigned. If a task changes at a given moment, drone swarm members first leave their respective task alliances, then randomly join the task at that moment, and again make decisions according to the alliance rules until none of the members meet the alliance rules at the time of decision-making, then the next task is assigned until all tasks are assigned. This ultimately forms a stable and efficient task alliance and resource allocation scheme.

[0108] Optimal task allocation scheme. Different alliance rules result in different alliance structures. The alliance rules set in this example are as follows:

[0109] The rules for joining an alliance are defined as follows:

[0110]

[0111] Among them, S p For the original alliance structure, S q For the new alliance structure, U m (S p U represents the total revenue of the original alliance structure. m (S q U represents the total revenue of the new alliance structure. n (S p (U) represents the revenue that drones gained within the original alliance. n (S q (U) refers to the benefits that drones will gain in the new alliance. g (S p ) represents the revenue that autonomous vehicles gained from the original alliance, U g (S q This refers to the benefits that autonomous vehicles will gain in the new alliance.

[0112] If a drone or unmanned vehicle does not contribute resources to the mission alliance, but the total revenue of the alliance structure increases and the revenue obtained by the drone or unmanned vehicle increases, then the drone or unmanned vehicle will join the mission.

[0113] The exit rule is defined as follows:

[0114]

[0115] Among them, S p For the original alliance structure, S q For the new alliance structure, U m (S p U represents the total revenue of the original alliance structure. m (S q U represents the total revenue of the new alliance structure. n (S p (U) represents the revenue that drones gained within the original alliance. n (S q (U) refers to the benefits that drones will gain in the new alliance. g (S p ) represents the revenue that autonomous vehicles gained from the original alliance, U g (S q This refers to the benefits that autonomous vehicles will gain in the new alliance.

[0116] If a drone or unmanned vehicle withdraws from a mission alliance, and the drone or unmanned vehicle's gains are not reduced, and the total gains of the alliance structure are not reduced, then the drone or unmanned vehicle will withdraw from the mission.

[0117] The transfer rules are defined as follows:

[0118]

[0119] Among them, S p For the original alliance structure, S q For the new alliance structure, U m (S p U represents the total revenue of the original alliance structure. m (S q ) represents the total revenue of the new alliance structure. For the benefit of the alliance currently responsible for the mission of the drone or unmanned vehicle in the original alliance structure. This refers to the revenue generated by transferring drones or unmanned vehicles to the original alliance structure. For the benefit of the alliance currently responsible for the mission of the drone or unmanned vehicle in the new alliance structure, Benefits of the mission alliance for the transfer of drones or unmanned vehicles to the new alliance structure.

[0120] Drones or unmanned vehicles from the Mission Alliance Transfer to the Mission Alliance At that time, the total benefit of the alliance structure increases, and the task alliance... The sum of the benefits is greater than in the Mission Alliance The benefits and profits are then transferred by drones or unmanned vehicles.

[0121] Based on the above rules, drones and unmanned vehicles make decisions until all tasks are assigned.

[0122] To verify the effectiveness of the method of the present invention, the following simulation experiment was conducted.

[0123] like Figure 3 The diagram shows a scenario after all tasks have been assigned in this invention. The application scenario involves 15 drones and 3 unmanned vehicles collaboratively completing 15 tasks. Different tasks have different resource requirements, and each drone swarm member carries different resources. Based on the resource requirements, the tasks can be categorized into four types: communication, reconnaissance, jamming, and coupled communication / reconnaissance.

[0124] Information such as the start and end times of the task, its value, priority, requirements, and resource attributes of the unmanned cluster members can be obtained from Table 1.

[0125] Table 1 - Scene Information Table

[0126]

[0127]

[0128] In Table 1, the task requirements, drone resources, and unmanned vehicle resources have been normalized.

[0129] Modify the start and end times of the mission, mission requirements, resources of drones and unmanned vehicles, and topology to conduct simulation analysis under different scenarios.

[0130] Figure 4 The paper presents the changes in cumulative returns of alliance structures formed using the overlapping alliance formation distributed collaborative scheduling method, random allocation algorithm, and sequential overlapping alliance formation algorithm of this invention. Compared with the random algorithm and sequential overlapping alliance game algorithm, the overlapping alliance formation distributed collaborative scheduling method of this invention increases the cumulative returns of the alliance structure by 67.2% and 15.6%, respectively.

[0131] Figure 5 The paper presents the total revenue per second of the alliance structure formed using the overlapping alliance formation distributed cooperative scheduling method, random allocation algorithm, and sequential overlapping alliance formation algorithm of this invention. For example, in the fourth second, compared with the random algorithm and the sequential overlapping alliance game algorithm, the overlapping alliance formation distributed cooperative scheduling method of this invention increases the cumulative revenue of the alliance structure by 78% and 26%, respectively.

[0132] Figure 6 The success rates of tasks using the overlapping alliance formation distributed cooperative scheduling method, random allocation algorithm, and sequential overlapping alliance formation algorithm of this invention are presented. Figure 6 This demonstrates that, compared to random allocation algorithms and sequential overlapping alliance algorithms, the overlapping alliance-based distributed collaborative scheduling method of this invention improves the task success rate.

[0133] Example 3

[0134] This invention provides a distributed cooperative scheduling device for unmanned clusters to perform multi-modal heterogeneous dynamic tasks, comprising:

[0135] Task dynamic scheduling module: In response to changes in tasks, it allocates multi-mode heterogeneous dynamic tasks according to preset dynamic task collaborative scheduling rules until all unmanned cluster members no longer want to change their own decisions to improve benefits.

[0136] The static task scheduling module is used to allocate multi-mode heterogeneous dynamic tasks according to preset static task collaborative scheduling rules when the task has not changed, until all unmanned cluster members no longer want to change their own decisions to improve their benefits.

[0137] Example 4

[0138] This invention provides a distributed cooperative scheduling system for unmanned clusters to perform multi-modal heterogeneous dynamic tasks, comprising:

[0139] Memory, used to store instructions;

[0140] The processor is configured to execute the instructions, causing the device to perform operations that implement a distributed cooperative scheduling method for unmanned clusters to perform multi-modal heterogeneous dynamic tasks as described in Embodiment 1:

[0141] In response to changes in tasks, multi-mode heterogeneous dynamic tasks are allocated according to preset dynamic task collaborative scheduling rules until all unmanned cluster members no longer want to change their own decisions to improve benefits.

[0142] In response to the unchanged task, multi-mode heterogeneous dynamic tasks are allocated according to the preset static task collaborative scheduling rules until all unmanned cluster members no longer want to change their own decisions to improve their benefits.

[0143] Example 5

[0144] This invention provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the operation of a distributed cooperative scheduling method for unmanned clusters to perform multi-modal heterogeneous dynamic tasks as described in Embodiment 1.

[0145] In response to changes in tasks, multi-mode heterogeneous dynamic tasks are allocated according to preset dynamic task collaborative scheduling rules until all unmanned cluster members no longer want to change their own decisions to improve benefits.

[0146] In response to the unchanged task, multi-mode heterogeneous dynamic tasks are allocated according to the preset static task collaborative scheduling rules until all unmanned cluster members no longer want to change their own decisions to improve their benefits.

[0147] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, this application can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this application can take the form of a computer program product embodied on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0148] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0149] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0150] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0151] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the technical principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A distributed cooperative scheduling method for unmanned swarms to perform multi-modal heterogeneous dynamic tasks, characterized in that, include: In response to changes in tasks, multi-mode heterogeneous dynamic tasks are allocated according to preset dynamic task collaborative scheduling rules until all unmanned cluster members no longer want to change their own decisions to improve benefits. In response to the fact that the task has not changed, the multi-mode heterogeneous dynamic task is allocated according to the preset static task collaborative scheduling rules until all unmanned cluster members no longer want to change their own decisions to improve the benefits. Before allocating multi-modal heterogeneous dynamic tasks according to dynamic / static task collaborative scheduling rules, the following steps are also included: Obtain the initial information of multi-mode heterogeneous dynamic tasks and unmanned cluster members, and randomly allocate the multi-mode heterogeneous dynamic tasks accordingly to obtain the alliance under the initial allocation scheme. Calculate the total revenue and average success rate of the alliance structure under the initial allocation scheme. The total revenue and average success rate of the alliance structure are used to provide data judgment basis for dynamic / static task collaborative scheduling rules. The total benefit of the alliance structure is the sum of the benefits of all alliances, the benefit of an alliance is the sum of the benefits of its members, and the benefit of each member is utility minus cost. The alliance's revenue is calculated using the following formula: Among them, U m (Task) represents the reward for the alliance to execute multi-mode heterogeneous dynamic task m, where n is the drone and g is the unmanned vehicle. The benefits for alliance member drones n performing multi-mode heterogeneous dynamic tasks m. The benefits for autonomous vehicles g performing multi-modal heterogeneous dynamic tasks m for alliance members. For a collection of UAVs in a multi-mode heterogeneous dynamic task m, For the set of unmanned vehicles in the multimodal heterogeneous dynamic task m; Among them, R n R g C n C g The utility and cost of performing multimodal heterogeneous dynamic tasks by alliance member drones n and unmanned vehicles g, respectively; The utility of alliance members is obtained through the following methods: Among them, V m For the value of the mission alliance m, F m Let ω1 be the priority of the mission alliance m, and D be the balance coefficient. n Assign resources to the mission alliance m for member drones, D g Allocate resources from mission alliance m to the autonomous vehicles of alliance members, S n and S g To improve the success rate of missions performed by member drones and unmanned vehicles; The success rate of alliance member drones and unmanned vehicles performing alliance missions can be obtained using the following methods: in, These represent the communication resource requirements, reconnaissance resource requirements, and jamming resource requirements for task m, respectively. n Assign resources to the mission alliance m for member drones, D g Allocate resources from mission alliance m to the autonomous vehicles of alliance members; The cost to alliance members is obtained through the following methods: Where β, λ1, λ2, and λ3 are balance coefficients. The communication power consumption, reconnaissance power consumption, and jamming power consumption of the alliance member UAV n performing mission m. The communication power consumption, reconnaissance power consumption, and jamming power consumption of the unmanned vehicle g performing mission m for alliance member; Among them, U m (S) represents the total revenue of the alliance structure, m represents the multi-modal heterogeneous dynamic task, and M represents the set of multi-modal heterogeneous dynamic tasks. The average success rate of the alliance structure is calculated using the following formula: Where Suc is the average success rate of the alliance structure, L is the number of the multimodal heterogeneous dynamic tasks, and suc m The success rate of executing the multimodal heterogeneous dynamic task m is given by M, where M is the set of multimodal heterogeneous dynamic tasks. The success rate of the multimodal heterogeneous dynamic task m is calculated using the following method: in, This represents the communication resources contributed by drone n, a member of the alliance, to mission m. This represents the reconnaissance resources contributed by member drone n to mission m. This represents the interference resources contributed by drone n, a member of the alliance, to mission m. This indicates the communication resources contributed by the autonomous vehicle g from the alliance member to task m. This indicates the reconnaissance resources contributed by the alliance member's unmanned vehicle g to mission m. This indicates the interference resources contributed by the autonomous vehicle g from the alliance member to task m. This indicates the communication relay resources contributed by the autonomous vehicle g from the alliance member to task m. This represents the reconnaissance computing resources contributed by the alliance member's unmanned vehicle g to mission m. These are the communication resource requirements, reconnaissance resource requirements, and jamming resource requirements for mission coalition m, respectively.

2. The distributed cooperative scheduling method for unmanned swarms to perform multi-modal heterogeneous dynamic tasks according to claim 1, characterized in that, The response to task changes, including allocating multi-modal heterogeneous dynamic tasks according to preset dynamic task collaborative scheduling rules, includes: In response to a change in tasks, retrieve the multi-modal heterogeneous dynamic tasks that occurred in both the current and previous time steps. Members of the alliances of these multi-modal heterogeneous dynamic tasks leave their respective alliances. Then, randomly assign the multi-modal heterogeneous dynamic tasks at the current time step and execute the following loop: The unmanned cluster members make decisions to obtain a new alliance structure. In response to the unmanned cluster members meeting the preset alliance change rules, the alliance structure is updated under the rules, and the total revenue and average success rate of the alliance structure are updated.

3. The distributed cooperative scheduling method for unmanned swarms to perform multi-modal heterogeneous dynamic tasks according to claim 1, characterized in that, The response, assuming no change in the task, involves allocating multi-modal heterogeneous dynamic tasks according to preset static task collaborative scheduling rules, including: If the task remains unchanged, execute the following loop: The unmanned cluster members make decisions to obtain a new alliance structure. In response to the unmanned cluster members meeting the preset alliance change rules, the alliance structure is updated under the rules, and the total revenue and average success rate of the alliance structure are updated.

4. A distributed cooperative scheduling method for unmanned swarms to perform multi-modal heterogeneous dynamic tasks according to any one of claims 2 or 3, characterized in that, The response to unmanned cluster members satisfying a preset alliance change rule, and the execution of an update to the alliance structure under that rule, includes: If joining a consortium increases the total revenue of the consortium and the revenue gained by the unmanned cluster member, then the unmanned cluster member joins the consortium. If the total revenue of the alliance structure and the revenue obtained by the unmanned cluster member do not decrease after the unmanned cluster member leaves the alliance, then the unmanned cluster member leaves the alliance. If the total revenue of the alliance structure increases after a member of an unmanned cluster moves from its current alliance to another alliance, and the sum of the revenues of the two alliances involved after the move is greater than the sum of the revenues of the two alliances involved before the move, then the member of the unmanned cluster moves from its current alliance to another alliance.

5. A distributed cooperative scheduling device for unmanned swarms to perform multi-modal heterogeneous dynamic tasks based on the method of any one of claims 1 to 4, characterized in that, include: Task dynamic scheduling module: In response to changes in tasks, it allocates multi-mode heterogeneous dynamic tasks according to preset dynamic task collaborative scheduling rules until all unmanned cluster members no longer want to change their own decisions to improve benefits. The static task scheduling module is used to allocate multi-mode heterogeneous dynamic tasks according to preset static task collaborative scheduling rules when the task has not changed, until all unmanned cluster members no longer want to change their own decisions to improve their benefits.

6. A distributed cooperative scheduling system for unmanned swarms to perform multi-modal heterogeneous dynamic tasks, characterized in that, include: Memory, used to store instructions; A processor is configured to execute the instructions, causing the system to perform operations that implement the distributed cooperative scheduling method for unmanned clusters to perform multi-modal heterogeneous dynamic tasks as described in any one of claims 1-4.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the operation of a distributed cooperative scheduling method for unmanned clusters to perform multi-modal heterogeneous dynamic tasks as described in any one of claims 1-4.

Images