Aircraft cluster task allocation method under direct connection condition of satellite and aircraft

Abstract

An aircraft cluster task allocation method under a satellite and aircraft direct connection condition utilizes a satellite to sense dynamic target situation information, downloads the dynamic target situation information to an aircraft cluster through a direct connection link, and adopts a dynamic task allocation algorithm based on local re-planning sequence greedy search. All aircrafts coordinate execution steps and adjust step ranges under a rolling distribution framework, idle computing resources and finite time windows of the aircrafts in the execution steps are utilized to generate a greedy redistribution scheme containing dynamic targets, and high-priority assignment of the dynamic targets is ensured on the basis of meeting online solving timeliness.

Description

Technical Field

[0001] The invention relates to an aircraft cluster task allocation method under the condition of direct connection between a satellite and an aircraft, which is used for realizing autonomous task allocation of aircraft facing dynamic targets. Background Art

[0002] Task allocation technology is to coordinate and plan between multiple tasks and aircraft clusters, taking into account constraints such as task requirements and aircraft performance, and reasonably assigning tasks to each aircraft with the goal of minimizing task costs or maximizing benefits. However, in the process of collaborative execution of tasks by aircraft clusters, information such as the number and location of aircraft and targets may change in real time, and the demand for the immediacy of obtaining high-dynamic and fast-changing scene situations and the timeliness of task allocation solutions is becoming more and more urgent. Summary of the invention

[0003] The technical problem to be solved by the present invention is to overcome the deficiencies of the prior art and obtain a more globally optimal task allocation solution with good real-time performance.

[0004] The purpose of the present invention is achieved through the following technical solutions:

[0005] A method for allocating tasks to a cluster of aircraft under the condition of direct connection between satellites and aircraft is proposed. The method uses satellites to perceive the situation information of dynamic targets and transmits it to the aircraft cluster through a direct link. A dynamic task allocation algorithm based on greedy search of local replanning sequence is adopted. Each aircraft coordinates the execution step and the adjustment step range under a rolling allocation framework. The idle computing resources of the aircraft in the execution step and the limited time window are used to generate a greedy reallocation scheme containing dynamic targets. High-priority assignment of dynamic targets is ensured on the basis of meeting the timeliness of online solution.

[0006] Compared with the prior art, the present invention has the following beneficial effects:

[0007] (1) Compared with the prior art, the method for allocating tasks of aircraft clusters under the condition of direct connection between satellites and aircraft proposed in the present invention has significantly improved the ability to perceive the dynamic target situation in global scenarios, and is more conducive to obtaining a more optimal task allocation plan globally.

[0008] (2) The dynamic task reallocation algorithm based on greedy search of local replanning sequence proposed in the present invention fully utilizes the idle computing resources in the aircraft during mission execution to generate a dynamic target reallocation plan, ensuring real-time solution while improving the overall benefit of dynamic allocation.

[0009] (3) The dynamic task reallocation algorithm based on greedy search of local replanning sequence proposed in the present invention designs a local target reallocation strategy, which effectively reduces the dimension of the allocation problem and reduces the time consumption of dynamic reallocation compared with traditional methods.

[0010] (4) The dynamic task reallocation algorithm based on greedy search of local replanning sequence proposed in the present invention adopts a rolling planning framework and has the ability to respond to dynamic targets in real time. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] Figure 1 Schematic diagram of the rolling planning process. DETAILED DESCRIPTION

[0012] To make the objectives, technical solutions and advantages of the present invention more clear, the embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0013] A method for allocating tasks to aircraft clusters under the condition of direct connection between satellites and aircraft is proposed. Through satellite perception of dynamic target situation information, a dynamic task allocation algorithm based on greedy search of local replanning sequence is proposed. Each aircraft coordinates the execution step and the adjustment step range under the rolling allocation framework, and uses the idle computing resources and limited time window of the aircraft in the execution step to generate a greedy reallocation scheme containing dynamic targets, ensuring the high priority assignment of dynamic targets on the basis of meeting the timeliness of online solution. The specific method includes:

[0014] (1) Satellite perception of dynamic targets

[0015] Aircraft clusters are limited by their detection distance and it is difficult to obtain the global target situation in real time. The traditional mode relies on the ground command center to comprehensively process the target situation information and upload it to the cluster, but the target information update cycle is long in this mode and is suitable for the pre-allocation stage. The present invention proposes a new mode of direct connection between satellites and aircraft, which generates target information on orbit and directly transmits it to the cluster, thereby improving the cluster's ability to recognize time-varying situations.

[0016] The specific steps are as follows:

[0017] (a) Generate dynamic target pointing information. Combining wide-area detection means and low-orbit intelligent remote sensing constellations, the satellite system continuously obtains the position, heading and speed information of dynamic targets in the specified area, and processes them on board to form pointing information. For newly discovered dynamic targets, the attributes of the new targets are marked in the pointing information; for targets that have been mastered, their basic information such as position, heading, speed, etc. is updated.

[0018] (b) Target pointing information is sent to the aircraft cluster. At fixed time intervals, the target pointing information in the specified area is sent to the aircraft cluster via the satellite and aircraft direct link. The aircraft cluster receives the target pointing information and updates the target situation map.

[0019] (2) Dynamic task allocation of aircraft clusters

[0020] Based on the target situation information obtained in real time, each aircraft coordinates the execution step and the adjustment step range under the rolling allocation framework, and uses the idle computing resources and limited time window of the aircraft in the execution step to generate a greedy reallocation plan that includes dynamic targets. The specific steps are as follows:

[0021] (a) Rolling planning framework. Before the execution of the aircraft cluster task, static task allocation is completed based on the target prior information. Each aircraft takes off and performs the coordinated approach and detailed inspection task of the designated area targets along the pre-set track. After obtaining the newly added dynamic target information transmitted by the satellite, the dynamic task reallocation process is triggered. Since the allocation solution is time-consuming, in order to ensure that the aircraft obtains the executable track in real time, the sequence of targets to be executed is divided into execution steps and adjustment steps. Each aircraft plans the flight track along the target sequence in the execution step, and releases the target sequence in the adjustment step, which together with the newly added dynamic targets constitute the dynamic allocation target set of this round. Each aircraft uses the idle computing resources in the execution step to complete the task reallocation, and modifies the adjustment step target execution sequence online according to the allocation result. When the newly added target information is obtained again, the above rolling planning process is repeated, such as Figure 1 shown.

[0022] The selection of execution step and adjustment step range affects the allocation performance. If each aircraft selects a larger execution step range, it provides a sufficient computational time window for the task allocation algorithm, but the remaining adjustable target range is reduced, the overall benefit of task redistribution is limited, and it is difficult to assign aircraft to quickly perform reconnaissance tasks of dynamic targets in a timely manner; if each aircraft selects a smaller execution step range, the adjustment target range increases, the overall task redistribution benefit increases, but the task redistribution algorithm solution time window is reduced. Therefore, on the basis of ensuring that dynamic targets are quickly assigned, it is necessary to design a local target redistribution strategy to reduce the dimension of the allocation problem and reduce the time consumption of dynamic redistribution.

[0023] (b) Execution step and adjustment step range selection mechanism. Calculate the remaining time for each aircraft to approach the current first target. If the minimum remaining time is greater than 5 seconds, the execution step only includes the current first target to be executed, and the remaining targets are all divided into the adjustment step range and participate in the redistribution process together; if the minimum remaining time is greater than 1 second and less than 5 seconds, the execution step of the three aircraft closest to the newly added dynamic target Dubins includes the current first target to be executed, and the remaining targets are divided into the adjustment step range; the adjustment step range of the aircraft farthest from the newly added dynamic target is empty, and the adjustment space of the remaining aircraft is added in order from far to near to the newly added dynamic target; if the minimum remaining time is greater than 0.3 seconds and less than 1 second, the execution step of the aircraft closest to the newly added dynamic target Dubins includes the current first target to be executed, and the adjustment step range of the remaining aircraft is empty. If the minimum remaining time is less than 0.3 seconds, it indicates that there is an aircraft in the system that happens to be close to the target to be executed, and its minimum remaining time can be postponed to the next target, and the above rules are repeated to determine the execution step range.

[0024] (c) Sequential greedy search algorithm based on local replanning. The targets within each aircraft adjustment step and the newly added dynamic targets together constitute the target set for this dynamic task allocation. A sequential greedy search algorithm based on local replanning is designed to realize the dynamic task allocation of aircraft clusters to dynamic targets. The specific steps of the algorithm are as follows.

[0025] Step 1: Calculate the initial allocation plan. Determine the target range for each aircraft in the cluster to participate in this round of dynamic reallocation based on the execution step and adjustment step range selection mechanism, and generate the initial allocation plan for each aircraft in this round of rolling planning cycle accordingly.

[0026]

[0027] In the formula, The target allocation sequence obtained during the rolling planning cycle for vehicle i, For sequence Intercepted front κ i elements, κ i The last target of the sequence executed by aircraft i is in the sequence The order in which the steps are performed is determined by the execution step and adjustment step range selection mechanism. The initial target execution sequence of aircraft i in this rolling planning cycle. A collection of all aircraft in the cluster.

[0028] Step 2: Parameter initialization. The set of aircraft participating in dynamic reallocation in this round The target set is

[0029]

[0030] In the formula, is the set of targets to be executed, T * Add a new dynamic target for the current one. It represents the target set that does not participate in dynamic task reallocation in this round, which is composed of the targets within the execution step of each aircraft.

[0031]

[0032] The length counter of the aircraft i target execution sequence is η i , the initial value is Sure.

[0033]

[0034] Step 3: Calculate target revenue. Each target is added to the initial target execution sequence of the aircraft in turn Calculate the target return.

[0035]

[0036] In the formula, It represents the benefit of vehicle i executing target j in the first iteration of the algorithm. The specific calculation method is as follows.

[0037]

[0038] In the formula, λ h <1 represents the attenuation coefficient of target h, represents the base value of the target h, Indicates that aircraft i executes the sequence along the initial target The time when target h is executed, is the time when target j appears. Operator Indicates adding a new element after the nth element in the sequence.

[0039] Step 4: Calculate the number of algorithm iterations. The algorithm exit condition is when All targets have been assigned or the execution sequence of each aircraft target has reached the upper limit. If the conditions are met, the algorithm exits. The upper limit of the number of algorithm iterations is N min The calculation method is

[0040]

[0041] In the formula, C i The upper limit of the number of target executions for vehicle i.

[0042] Step 5: Sequential greedy search and parameter update. The current number of iterations is n, and the initial value of n is 1. According to the greedy principle, select the allocation pair that currently produces the maximum target benefit. As the allocation scheme for the nth round,

[0043]

[0044] In the formula, represents the benefit of aircraft i executing target j in the nth iteration of the algorithm, Meaning: Get the values ​​of i and j corresponding to the maximum value. The range of values ​​of i and j is

[0045] Update the parameters of the aircraft The target execution sequence length counter is incremented by 1.

[0046]

[0047] Will From the target collection Delete it.

[0048]

[0049] In the aircraft Insert at the end of the target execution sequence

[0050]

[0051] In the formula, the operator Indicates adding a new element after the last element of the sequence.

[0052] If the aircraft Reached the upper limit of the target execution quantity Then From the aircraft collection Delete the corresponding bid value, and clear it to zero. No longer participate in the subsequent allocation process.

[0053]

[0054] Step 6: Update target revenue. The target return of the n+1th round is updated sequentially, that is,

[0055]

[0056] Return to step 5, when n reaches the upper limit of the number of algorithm iterations N min , proceed to step 7.

[0057] Step 7: Allocation scheme integration and storage. After the algorithm cycle is completed, the target sequence assigned to each aircraft is Integrate and store to provide input information for dynamic task reallocation in the next rolling cycle.

[0058] The present invention proposes a method for allocating tasks for aircraft clusters under the condition of direct connection between satellites and aircraft. The method combines the advantages of space-based wide-area situational awareness and patrol and close-range detailed inspection, can quickly respond to the dynamic global environment, and customizes a dynamic task allocation algorithm based on greedy search of local replanning sequences. The method ensures the rapid assignment of high-value dynamic targets on the basis of meeting the timeliness of online solution, thereby improving the reliability and optimality of the dynamic task allocation algorithm for aircraft clusters.

[0059] The contents not described in detail in the specification of the present invention belong to the common knowledge of those skilled in the art.

[0060] Although the present invention has been disclosed as above in the form of a preferred embodiment, it is not intended to limit the present invention. Any person skilled in the art may make possible changes and modifications to the technical solution of the present invention by using the methods and technical contents disclosed above without departing from the spirit and scope of the present invention. Therefore, any simple modifications, equivalent changes and modifications made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall fall within the protection scope of the technical solution of the present invention.

Claims

1. A method for allocating tasks of aircraft clusters under the condition of direct connection between satellites and aircraft, characterized in that: include: Generate dynamic target pointing information and send it to the aircraft cluster via satellite; Before the execution of the aircraft cluster mission, static task allocation is completed according to the target prior information; each aircraft takes off and executes the coordinated approach mission of the designated area target along the pre-set track, and triggers the dynamic task reallocation process after obtaining the newly added dynamic target indication information transmitted by the satellite; according to the target situation information obtained in real time, each aircraft coordinates the execution step and the adjustment step range under the rolling allocation framework, and uses the idle computing resources of the aircraft in the execution step and the limited time window to generate a greedy reallocation plan including dynamic targets.

2. The method for allocating tasks of aircraft clusters according to claim 1, characterized in that: The rolling allocation framework refers to obtaining the newly added dynamic target information transmitted by the satellite, and dividing the target sequence to be executed into execution steps and adjustment steps to ensure that the aircraft obtains the executable track in real time; Each aircraft plans its flight trajectory along the target sequence in the execution step and releases the target sequence in the adjustment step, which together with the newly added dynamic targets constitute the target set of this round of dynamic allocation; each aircraft uses the idle computing resources in the execution step to complete task reallocation, and modifies the adjustment step target execution sequence online according to the allocation results; when the newly added target information is obtained again, the above rolling planning process is repeated.

3. The method for allocating tasks of aircraft clusters according to claim 1, characterized in that: The selection mechanism of the execution step and adjustment step range is as follows: calculate the remaining time for each aircraft to approach the current first target. If the minimum remaining time is greater than 5 seconds, the execution step only includes the current first target to be executed, and the remaining targets are all divided into the adjustment step range and participate in the redistribution process in a unified manner; if the minimum remaining time is greater than 1 second and less than 5 seconds, the execution step of the three aircraft closest to the newly added dynamic target includes the current first target to be executed, and the remaining targets are divided into the adjustment step range; the adjustment step range of the aircraft farthest from the newly added dynamic target is empty, and the adjustment space of the remaining aircraft is added in order from far to near to the newly added dynamic target; if the minimum remaining time is greater than 0.3 seconds and less than 1 second, the execution step of the aircraft closest to the newly added dynamic target includes the current first target to be executed, and the adjustment step range of the remaining aircraft is empty; if the minimum remaining time is less than 0.3 seconds, it indicates that there is an aircraft in the system that happens to be close to the target to be executed, and its minimum remaining time is postponed to the next target, and the above rules are repeated to determine the execution step range.

4. The method for allocating tasks of aircraft clusters according to claim 1, characterized in that: A greedy reallocation scheme is obtained by using a sequential greedy search algorithm based on local replanning, which includes: Step 1: Calculate the initial allocation plan; Step 2: Parameter initialization; Step 3: Calculate target return; Step 4: Calculate the number of algorithm iterations; Step 5: Update sequence greedy search and parameters; Step 6: Update the target profit. If the algorithm reaches the upper limit of iterations, go to step 7, otherwise go to step 5. Step 7: Allocation plan integration and storage.

5. The method for allocating tasks of aircraft clusters according to claim 4, characterized in that: The initial allocation plan is as follows: In the formula, The target allocation sequence obtained during the rolling planning cycle for vehicle i, For sequence Intercepted front κ i elements, κ i The last target of the sequence executed by aircraft i is in the sequence The order in is determined by the execution step and adjustment step range selection mechanism; The initial target execution sequence of aircraft i in this rolling planning cycle; A collection of all aircraft in the cluster.

6. The method for allocating tasks of aircraft clusters according to claim 5, characterized in that: Parameter initialization specifically includes: The set of aircraft participating in dynamic reallocation in this round The target set is In the formula, is the set of targets to be executed, T * Add a new dynamic target for the current one. It represents the target set that does not participate in the dynamic task reallocation in this round, which is composed of the targets within the execution step of each aircraft; The length counter of the aircraft i target execution sequence is η i , the initial value is Sure; 7. The method for allocating tasks of aircraft clusters according to claim 6, characterized in that: The target return calculation includes: Will Each target is added to the initial target execution sequence of the aircraft in turn Calculate target returns; In the formula, It represents the benefit of vehicle i executing target j in the first iteration of the algorithm. The specific calculation method is as follows: In the formula, λ h <1 represents the attenuation coefficient of target h, represents the base value of the target h, Indicates that aircraft i executes the sequence along the initial target The time when target h is executed, is the time when target j appears, operator Indicates adding a new element after the nth element in the sequence.

8. The method for allocating tasks of aircraft clusters according to claim 7, characterized in that: The calculation algorithm iteration number specifically includes: The algorithm exit condition is when All targets have been assigned or the execution sequence of each aircraft target has reached the upper limit. If the conditions are met, the algorithm exits; the upper limit of the number of algorithm iterations is N min The calculation method is In the formula, C i The upper limit of the number of target executions for vehicle i.

9. The method for allocating tasks of aircraft clusters according to claim 8, characterized in that: Sequence greedy search and parameter update specifically include: The current number of iterations is n, and the initial value of n is 1; according to the greedy principle, the allocation pair that currently produces the maximum target benefit is selected As the allocation scheme for the nth round, In the formula, represents the benefit of aircraft i executing target j in the nth iteration of the algorithm, Meaning: Get the values ​​of i and j corresponding to the maximum value. The range of values ​​of i and j is Update the parameters of the aircraft The target execution sequence length counter is incremented by 1; Will From the target collection Delete from; In the aircraft Insert at the end of the target execution sequence In the formula, the operator Indicates adding a new element after the last element of the sequence; If the aircraft Reached the upper limit of the target execution quantity Then From the aircraft collection Delete the corresponding bid value, and clear it to zero. No longer participate in the subsequent allocation process; 10. The method for allocating tasks of aircraft clusters according to claim 1, characterized in that: The target revenue update specifically includes: According to the current The target return of the n+1th round is updated sequentially, that is,

Images