Satellite mission planning method and system for contract network based on concurrency mechanism

Abstract

The present invention provides a satellite mission planning method, system, storage medium and electronic device based on a concurrent mechanism for a contract net, which relates to the field of satellite scheduling. According to a preset mathematical model of a slave satellite, each slave satellite generates a corresponding current bidding scheme for the current unarranged task sequence by using the greedy rule with the maximum benefit; according to a preset mathematical model of a master satellite, the master satellite uses a multi-attribute bid evaluation strategy for bid evaluation twice before and after, selects a winning bid scheme from each of the bidding schemes, and updates the planning schemes of the first and second winning bid slave satellites; and finally outputs a globally optimal satellite mission planning scheme. The proposed concurrent mechanism includes full-task bidding and a secondary bid evaluation strategy, which reduces the communication volume generated by the contract net protocol during the bidding process by introducing the concurrent mechanism, and overcomes the problem of large communication volume in traditional contract net negotiation; and improves the bid evaluation criteria for using the contract net to solve the multi-satellite and multi-task allocation problem through the multi-attribute bid evaluation strategy.

Description

Technical Field

[0001] The present invention relates to the field of satellite scheduling, and specifically to a satellite task planning method, system, storage medium and electronic device based on a contract net with a concurrency mechanism. Background Art

[0002] With the continuous development of satellite technology in China, the role played by satellites is becoming increasingly important. Among them, remote sensing satellites can quickly and accurately obtain surface information, have a wide range of applications, and have become an important symbol of a country's comprehensive strength. At the same time, they also play an important role in fields such as environmental disaster prevention and control, urban construction planning, and weather forecasting.

[0003] Research on satellite scheduling problems at home and abroad focuses on multi-satellite and multi-task scheduling. Traditional multi-satellite and multi-task planning problems are mostly processed using a centralized architecture based on ground offline planning, but the centralized planning has many defects such as poor task response ability and low execution efficiency. The distributed architecture has become the main architecture method for current multi-satellite task planning due to its advantages such as decentralized communication and good robustness. The contract net is an effective method for solving distributed task planning and has been widely used in distributed satellite planning due to its high allocation efficiency and strong adaptability to dynamic environments.

[0004] The contract net is a classic negotiation strategy proposed for task and resource allocation. The contract net divides the member roles in the system into managers and contractors and realizes task allocation by imitating the "bidding, winning the bid" mechanism in economic behavior. Agents cooperate with each other and compete for tasks through bid values, pursuing global optimality on a locally optimal system configuration, so as to complete tasks with the optimal system configuration and the lowest cost. The contract net protocol is a classic and effective method for solving distributed satellite task allocation, and its main work process is: the main satellite issues a task, each slave satellite bids for the task, and the main satellite selects a winning bid plan from each bidding plan and allocates the task to the winning bid satellite.

[0005] However, the defects of the contract net protocol are also relatively obvious, especially the relatively large negotiation communication volume. Most traditional contract nets adopt the mode of "single task bidding and single agent winning the bid", which will generate a large amount of communication. Summary of the Invention

[0006] (1) Technical Problems to be Solved

[0007] Aiming at the deficiencies of the prior art, the present invention provides a satellite task planning method, system, storage medium and electronic device based on a contract net with a concurrency mechanism, and solves the technical problem of generating a relatively large negotiation communication volume in the process of obtaining a satellite task planning scheme by the traditional contract net.

[0008] (2) Technical Solutions

[0009] To achieve the above object, the present invention is realized by the following technical solutions:

[0010] A satellite mission planning method for a contract network based on a concurrency mechanism, including a number of slave satellites and a master satellite for decision-making. The method includes:

[0011] S1. The master satellite obtains the to-be-observed task sequence, enters the cyclic tendering process, sets q = 1, and sets the cyclic termination times.

[0012] S2. The master satellite starts the q-th tendering, obtains the currently unarranged task sequence. If the currently unarranged task sequence is empty, go to S6; otherwise, go to S3.

[0013] S3. The master satellite broadcasts the currently unarranged task sequence to each of the slave satellites; according to the preset slave satellite mathematical model, each of the slave satellites generates a corresponding current tendering plan for the currently unarranged task sequence by using the greedy rule of maximizing the benefit. If all the tendering plans are empty, go to S6; otherwise, go to S4.

[0014] S4. According to the preset master satellite mathematical model, the master satellite uses a multi-attribute bid evaluation strategy for bid evaluation, selects the winning bid plan from each of the tendering plans, updates the planning plan of the first winning slave satellite, and updates the currently unarranged task sequence for the first time. If the currently unarranged task sequence after this update is empty, go to S6; otherwise, go to S5.

[0015] S5. The master satellite uses the multi-attribute bid evaluation strategy for bid evaluation again, selects the winning bid plan from each of the tendering plans for the currently unarranged task sequence after the first update, updates the planning plan of the second winning slave satellite, and updates the currently unarranged task sequence for the second time; the multi-attribute bid evaluation strategy at least includes the total observation benefit of the satellite planning tasks.

[0016] S6. If all tasks have been arranged for observation, or there is no tendering plan, or the total observation benefit reaches the same cyclic termination times continuously, the planning ends, and the globally optimal satellite mission planning plan is output; otherwise, set q = q + 1, and go to S2.

[0017] Preferably, the slave satellite mathematical model in S3 includes an objective function of maximizing the benefit:

[0018]

[0019] And a first constraint condition:

[0020]

[0021]

[0022] OTS ij +dur j =OTE ij (4)

[0023]

[0024] e ij =e i dur j , i ∈ {1, …, m} (6)

[0025]

[0026] Among them, formula (2) represents the uniqueness constraint, that is, a task can be observed at most once; formula (3) represents that the time window requirement must be met when observing a task; formula (4) represents the relationship between the actual observation end time of a task and the task observation duration; formulas (5 - 6) represent the energy constraint and the energy consumption calculation method in satellite planning; formula (7) represents the storage constraint;

[0027] The main satellite S′, the set of slave satellites S = {S1, S2, …, S i , …, S m}, there are m slave satellites in total;

[0028] T = {t1, t2, …, t j , …,.t n} represents the sequence of tasks to be observed, and there are n tasks in total;

[0029] t j = <P j , dur j , dt j >, that is, the attributes of the task are represented by a triple, where P j represents the observation benefit of task t j , dur j represents the observation duration of task t j , dt j represents the task observation deadline of task t j ;

[0030] represents the set of visible time windows, indicating the set of visible time windows of task t j on slave satellite S i , NTM ij represents the number of time windows;

[0031] That is, the visible time window is represented by a binary tuple, indicating task tj At the k-th visible time window from satellite S i , and represent the start time and end time of the visible time window respectively;

[0032] OTW ij = <OTS ij , OTE ij >, that is, the observation time window is represented by a binary tuple. OTW ij represents the actual observation time window of mission t j from satellite S i . OTS ij and OTE ij represent the start time and end time of the observation time window respectively;

[0033] W p and W r represent the weights of the observation benefit and the perturbation respectively;

[0034] e i represents the unit energy consumption of the observation mission from satellite S i ;

[0035] e ij represents the energy consumed by the observation mission t i from satellite S j ;

[0036] E i represents the maximum energy limit of satellite S i ;

[0037] c ij represents the storage amount consumed by the observation mission t i from satellite S j ;

[0038] C i represents the maximum storage limit of satellite S i ;

[0039] represents whether mission t j is observed at the k-th visible time window from satellite S i . If observed, the value is 1, otherwise 0.

[0040] Preferably, the main satellite mathematical model in S5 includes

[0041] Objective function one for maximizing the total observation benefit:

[0042]

[0043] Objective function two for minimizing the total energy consumed:

[0044]

[0045] Objective function three for minimizing the mission completion time of the satellite planning scheme:

[0046]

[0047] And the second constraint condition:

[0048]

[0049] ET iq ≤PTE, i ∈ {1, …, m}, q ∈ {1, 2, …} (12)

[0050] Among them, formula (9) represents the constraint during the bidding process between the number of tasks; formula (10) represents the planning time deadline constraint; formula (11) represents the planning time deadline constraint;

[0051] ET iq represents the observation end time of the scheme generated for the task sequence T i from satellite S q in the q-th bidding, and where lst i represents the observation end time of the last task from satellite S i ; PTE represents the task planning deadline.

[0052] Preferably, the main satellite in S4 uses a multi-attribute evaluation strategy for bid evaluation and selects the winning bid from each of the bid proposals, specifically including:

[0053] S10. Use the AHP subjective weighting method to determine the weight of each preset objective, and the objectives include the total observation revenue, total energy consumed, and mission completion time of the satellite planning scheme;

[0054] S20. For each of the bid proposals, normalize the values of its respective objectives in a vector normalization manner to obtain a normalized decision matrix;

[0055] S30. According to the normalized decision matrix, use the TOPSIS bid evaluation method to obtain the final index values of each of the bid proposals;

[0056] S40. Take the bid proposal with the highest final index value as the winning bid.

[0057] Preferably, using the AHP subjective weighting method to determine the weight of each preset objective in S10 specifically includes:

[0058] S101. Construct a judgment matrix according to each of the said objectives and a preset objective quantification value standard;

[0059] S102. Obtain the weight of each of the said objectives according to the judgment matrix.

[0060] Preferably, the said S30 specifically includes:

[0061] S301. According to the said specification decision matrix [z gk f×3 , obtain the ideal solution and the negative ideal solution

[0062]

[0063]

[0064]

[0065] where g represents the g-th bidding plan in this bidding, and there are f bidding plans in total; k represents the k-th objective, and there are 3 objectives in total; c gk represents each objective corresponding to the g-th bidding plan in matrix C g , which are the total observation revenue, the total energy consumed, and the task completion time of the satellite planning plan respectively;

[0066] S302. Calculate the weighted distances and between each of the said bidding plans and the ideal solution and

[0067]

[0068]

[0069] where w k represents the weight value of the K-th objective;

[0070] S303. Obtain the comprehensive evaluation index of each plan according to the said weighted distances and ;

[0071]

[0072] where CE g represents the comprehensive evaluation index of the g-th bidding plan.

[0073] ​Preferably, after the main satellite obtains the to-be-observed task sequence in S1, the to-be-observed task sequence is re-sorted in descending order of benefit; if there are multiple tasks with the same benefit, a secondary sorting is performed in ascending order of the task observation duration, and then it enters the cyclic bidding process.

[0074] A satellite task planning system based on a concurrent mechanism contract net, including a number of slave satellites and a main satellite for decision-making. The system includes:

[0075] A task acquisition module, used to execute S1, the main satellite obtains the to-be-observed task sequence, enters the cyclic bidding process, sets q = 1, and sets the cyclic termination number;

[0076] A bidding start module, used to execute S2, the main satellite starts the q-th bidding, obtains the currently unarranged task sequence. If the currently unarranged task sequence is empty, transfer to the scheme confirmation module to execute S6; otherwise, transfer to the scheme generation module S3;

[0077] A scheme generation module, used to execute S3, the main satellite broadcasts the currently unarranged task sequence to each slave satellite; according to the preset slave satellite mathematical model, each slave satellite generates a corresponding current bidding scheme using the greedy rule with the maximum benefit for the currently unarranged task sequence. If all the bidding schemes are empty, transfer to the scheme confirmation module to execute S6; otherwise, transfer to the first scheme evaluation module to execute S4;

[0078] A first scheme evaluation module, used to execute S4, according to the preset main satellite mathematical model, the main satellite uses a multi-attribute evaluation strategy for evaluation, selects the winning scheme from each bidding scheme, updates the planning scheme of the first winning slave satellite, and updates the currently unarranged task sequence for the first time. If the currently unarranged task sequence after this update is empty, transfer to the scheme confirmation module to execute S6; otherwise, transfer to the second scheme evaluation module to execute S5;

[0079] A second scheme evaluation module, used to execute S5, the main satellite uses the multi-attribute evaluation strategy again for evaluation, for the currently unarranged task sequence after the first update, selects the winning scheme from each bidding scheme, updates the planning scheme of the second winning slave satellite, and updates the currently unarranged task sequence for the second time; the multi-attribute evaluation strategy at least includes the total observation benefit of the satellite planning task.

[0080] A scheme confirmation module, used to execute S6, if all tasks have been arranged for observation, or there is no bidding scheme, or the total observation benefit reaches the same cyclic termination number continuously, then the planning ends, and the globally optimal satellite task planning scheme is output; otherwise, set q = q + 1, and transfer to the bidding start module to execute S2.

[0081] A storage medium stores a computer program for satellite mission planning based on the contract net of the concurrency mechanism, wherein the computer program enables a computer to execute the satellite mission planning method as described above.

[0082] An electronic device includes:

[0083] One or more processors;

[0084] A memory; and

[0085] One or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the programs include those for executing the satellite mission planning method as described above.

[0086] (III) Beneficial effects

[0087] The present invention provides a satellite mission planning method, system, storage medium and electronic device based on the contract net of the concurrency mechanism. Compared with the prior art, the following beneficial effects are achieved:

[0088] In the present invention, according to the preset slave satellite mathematical model, each slave satellite generates a corresponding current bidding plan for the task sequence that has not been arranged yet by using the greedy rule with the maximum benefit; according to the preset master satellite mathematical model, the master satellite conducts bid evaluation twice before and after by using the multi-attribute bid evaluation strategy, selects the winning bid plan from each of the bidding plans, and updates the planning plans of the first and second winning bid slave satellites; and finally outputs a globally optimal satellite mission planning plan. The concurrency mechanism proposed by the present invention includes full task bidding and the secondary bid evaluation strategy. By introducing the concurrency mechanism, the communication volume generated in the bidding process of the contract net protocol is reduced, and the problem of large communication volume in the traditional contract net negotiation is overcome; in addition, the bid evaluation criteria for using the contract net to solve the multi-satellite and multi-task allocation problem are improved through the multi-attribute bid evaluation strategy. BRIEF DESCRIPTION OF THE DRAWINGS

[0089] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0090] Figure 1 It is a schematic flowchart of a satellite mission planning method based on the contract net of the concurrency mechanism provided by an embodiment of the present invention;

[0091] Figure 2Schematic diagram of a bid plan screening method provided by an embodiment of the present invention;

[0092] Figure 3 Block diagram of a satellite mission planning system of a contract net based on a concurrency mechanism provided by an embodiment of the present invention. Detailed implementation manners

[0093] To make the objectives, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention are described clearly and completely. Apparently, the described embodiments are some but not all of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0094] By providing a satellite mission planning method, system, storage medium and electronic device of a contract net based on a concurrency mechanism, the embodiments of the present application solve the technical problem of large negotiation communication volume generated in the process of obtaining a satellite mission planning plan by a traditional contract net.

[0095] The general idea of the technical solutions in the embodiments of the present application to solve the above technical problems is as follows:

[0096] In the embodiments of the present invention, according to a preset mathematical model of a slave satellite, each slave satellite generates a corresponding current bid plan for the current unarranged task sequence by using a greedy rule with the maximum benefit; according to a preset mathematical model of a master satellite, the master satellite performs bid evaluation twice before and after by using a multi-attribute bid evaluation strategy, selects a winning bid plan from each of the bid plans, and updates the planning plans of the first and second winning bid slave satellites; and finally outputs a globally optimal satellite mission planning plan. The concurrency mechanism proposed in the embodiments of the present invention includes full-task bidding and a secondary bid evaluation strategy. By introducing the concurrency mechanism, the communication volume generated by the contract net protocol in the bidding process is reduced, and the problem of large negotiation communication volume of the traditional contract net is overcome; in addition, the bid evaluation criteria for using the contract net to solve the multi-satellite multi-task allocation problem are improved by the multi-attribute bid evaluation strategy.

[0097] To better understand the above technical solutions, the above technical solutions will be described in detail below in conjunction with the accompanying drawings of the specification and specific implementation manners.

[0098] Embodiment:

[0099] First aspect, as Figure 1 shown, the embodiments of the present invention provide a satellite mission planning method of a contract net based on a concurrency mechanism, including a plurality of slave satellites and a master satellite for decision-making. The method includes:

[0100] S1. The master satellite obtains the task sequence to be observed, enters the loop tendering process, sets q = 1, and sets the loop termination count.

[0101] S2. The master satellite starts the q-th tendering, obtains the task sequence that has not been arranged currently. If the task sequence that has not been arranged currently is empty, go to S6; otherwise, go to S3.

[0102] S3. The master satellite broadcasts the task sequence that has not been arranged currently to each slave satellite; according to the preset mathematical model of the slave satellite, each slave satellite generates the corresponding current tendering plan using the greedy rule with the maximum benefit for this task sequence that has not been arranged currently. If all the tendering plans are empty, go to S6; otherwise, go to S4.

[0103] S4. According to the preset mathematical model of the master satellite, the master satellite uses the multi-attribute bid evaluation strategy for bid evaluation, selects the winning bid plan from each tendering plan, updates the planning plan of the first winning slave satellite, and updates the task sequence that has not been arranged currently for the first time. If the task sequence that has not been arranged currently after this update is empty, go to S6; otherwise, go to S5.

[0104] S5. The master satellite uses the multi-attribute bid evaluation strategy for bid evaluation again, selects the winning bid plan from each tendering plan for the task sequence that has not been arranged currently after the first update, updates the planning plan of the second winning slave satellite, and updates the task sequence that has not been arranged currently for the second time; the multi-attribute bid evaluation strategy includes at least the total observation benefit of the satellite planning task.

[0105] S6. If all tasks have been arranged for observation, or there is no tendering plan, or the total observation benefit reaches the same loop termination count continuously, the planning ends, and the globally optimal satellite task planning plan is output; otherwise, set q = q + 1, and go to S2.

[0106] The concurrency mechanism proposed in the embodiment of the present invention includes full-task tendering and secondary bid evaluation strategy. By introducing the concurrency mechanism, the communication volume generated in the tendering process of the contract net protocol is reduced, and the problem of large communication volume in the traditional contract net negotiation is overcome; in addition, the bid evaluation criteria for using the contract net to solve the multi-satellite and multi-task allocation problem are improved through the multi-attribute bid evaluation strategy.

[0107] The following will introduce each step of the above technical solution in detail in combination with specific content:

[0108] First of all, it should be noted that the "single-task tendering, single-agent winning bid" mode adopted by the traditional contract net will generate a large negotiation communication volume. The embodiment of the present invention designs a contract net based on the concurrency mechanism. The concurrency mechanism includes full-task tendering and secondary bid evaluation strategy, which is mainly used to solve the problem of large negotiation communication volume of the traditional contract net.

[0109] Full - task bidding means that in each bidding process, the main satellite will obtain the current un - scheduled observation task sequence and use all the un - scheduled tasks as bidding information for bidding. Secondary winning the bid means that in one bidding, first, a winning bid plan is selected (the screening of the bidding plan is as shown in Figure 2 ). Then, the un - scheduled observation task sequence is updated, and another winning bid plan is selected from the remaining bidding plans for the currently un - observed tasks.

[0110] The full - task bidding strategy can reduce the number of negotiations; in each round of the bidding process, the main satellite will update the current un - scheduled observation task sequence and re - bid it as bidding information. Each slave satellite will screen the bidding plan to decide whether to bid. The main steps are as follows:

[0111] S1. The main satellite obtains the task sequence to be observed, enters the loop bidding process, sets q = 1, and sets the loop termination times.

[0112] After the main satellite obtains the task sequence to be observed, it re - sorts the task sequence to be observed from the largest to the smallest according to the benefits; if there are multiple tasks with the same benefit, it performs a secondary sort from the smallest to the largest according to the task observation duration, and then enters the loop bidding process.

[0113] S2. The main satellite starts the q - th bidding, obtains the current un - scheduled task sequence. If the current un - scheduled task sequence is empty, go to S6; otherwise, go to S3.

[0114] S3. The main satellite broadcasts the current un - scheduled task sequence to each slave satellite; according to the preset slave - satellite mathematical model, each slave satellite generates the corresponding current bidding plan by using the greedy rule of the largest benefit for the current un - scheduled task sequence. If all the bidding plans are empty, go to S6; otherwise, go to S4.

[0115] In the prior art, the bid - evaluation strategy is not perfect. In the bidding process, the problem manager needs to uniformly evaluate the bids of all problem solvers to find the best problem solver. However, there are sometimes cases where bidders have the same ability, and in practice, lottery is usually used to decide. Then, all bidders, including the winning bidders and non - winning bidders, need to be replied to, which increases the system redundant communication overhead. In view of this, the embodiment of the present invention introduces a multi - attribute bid - evaluation strategy to solve the problem of imperfect bid - evaluation strategy in the traditional contract - net protocol. The multi - attribute bid - evaluation includes the AHP subjective weighting method and the TOPSIS bid - evaluation method.

[0116] After obtaining the tender information from each slave satellite, a corresponding bidding plan needs to be generated. In this step, a greedy-based adaptive simulated annealing contract net algorithm is designed to handle the autonomous task allocation and planning problem on each slave satellite. The algorithm encoding method uses integer encoding, and the specific content is as follows:

[0117] The slave satellite mathematical model in S3 includes an objective function for maximizing revenue:

[0118]

[0119] And the first constraint condition:

[0120]

[0121]

[0122] OTS ij +dur j =OTE ij (4)

[0123]

[0124] e ij =e i dur j ,i∈{1,…,m} (6)

[0125]

[0126] Among them, formula (2) represents the uniqueness constraint, that is, a task can be observed at most once; formula (3) means that the time window requirement must be met when observing a task; formula (4) represents the relationship between the actual observation end time of a task and the task observation duration; formulas (5 - 6) represent the energy constraint and the energy consumption calculation method in satellite planning; formula (7) represents the storage constraint;

[0127] The master satellite S′, the slave satellite set S = {S1, S2, …, S i ,…,S m}, there are m slave satellites in total;

[0128] T = {t1, t2, …, t j ,…,.t n} represents the sequence of tasks to be observed, and there are n tasks in total;

[0129] t j =<P j ,dur j ,dt j >, that is, the attributes of the task are represented by a triple, where P j represents the task tj Observation benefit, dur j Indicates task t j Observation duration, dt j Indicates task t j Observation deadline of the task;

[0130] Indicates the set of visible time windows, indicating task t j On satellite S i Set of visible time windows on, NTM ij Indicates the number of time windows;

[0131] That is, the visible time window is represented by a binary tuple, Indicates task t j On satellite S i The k-th visible time window on, And Respectively represent the start time and end time of this visible time window;

[0132] OTW ij = <OTS ij , OTE ij (>, that is, the observation time window is represented by a binary tuple, OTW ij Indicates task t j On satellite S i The actual observation time window on, OTS ij And OTE ij Respectively represent the start time and end time of this observation time window;

[0133] W p And W r Respectively represent the weights of the observation benefit and the perturbation;

[0134] e i Indicates the unit energy consumption for observing the task from satellite S i ;

[0135] e ij Indicates the energy consumed for observing task t from satellite S i ; j ;

[0136] E i Indicates the maximum energy limit of satellite S i ;

[0137] c ij Indicates the storage amount consumed for observing task t from satellite S i ; j ;

[0138] Ci Indicates from satellite S i 's maximum storage limit;

[0139] Indicates mission t j Whether to perform observations during the k-th visible time window of satellite S i If so, the value is 1; otherwise, it is 0.

[0140] S4. According to the preset main satellite mathematical model, the main satellite uses a multi-attribute bid evaluation strategy to evaluate bids, selects the winning bid from each of the bid proposals, updates the planning scheme of the first winning sub-satellite, and updates the task sequence that has not been arranged for the first time. If the task sequence that has not been arranged after this update is empty, go to S6; otherwise, go to S5;

[0141] The main satellite mathematical model includes

[0142] Objective function one for maximizing the total observation benefit:

[0143]

[0144] Objective function two for minimizing the total energy consumed:

[0145]

[0146] Objective function three for minimizing the task completion time of the satellite planning scheme:

[0147]

[0148] And the second constraint condition:

[0149]

[0150] ET iq ≤PTE, i∈{1,…,m}, q∈{1,2,…} (12)

[0151] Among them, formula (9) represents the constraint between the number of tasks during the bidding process; formula (10) represents the planning time deadline constraint; formula (11) represents the planning time deadline constraint;

[0152] ET iq Indicates that in the q-th bidding, for satellite S i For the task sequence T q The observation end time of the generated scheme, and Where lst i Indicates the observation end time of the last task on satellite S i ; PTE represents the task planning deadline.

[0153] In this step, the evaluation criteria for using the contract net to solve the multi-satellite multi-task allocation problem are improved through a multi-attribute evaluation strategy. Specifically, in S4, the main satellite uses a multi-attribute evaluation strategy to evaluate bids and select the winning bid from each of the bid proposals, which specifically includes:

[0154] S10. Use the AHP subjective weighting method to determine the weight of each preset objective. The objectives include the total observation benefit, total energy consumption, and task completion time of the satellite planning scheme, including:

[0155] S101. Construct a judgment matrix according to each of the objectives and the preset objective quantification value criteria.

[0156] Here, the relevant content of the judgment matrix is first supplemented. Assuming that the weight of objective a is w a , then represents the relative importance of objective a and objective b.

[0157] By comparing the importance of different objectives pairwise, estimate w ab , and form a judgment matrix A = [w ab 3×3. The value of w ab is shown in Table 1:

[0158] Table 1 Objective quantification value criteria

[0159] Target a relative to target b Quantification value Equally important 1 Slightly important 3 Important 5 Significantly important 7 Absolutely important 9 Intermediate value between two adjacent judgments 2，4，6，8

[0160] Through the above method, the judgment matrix A for the three objectives of the total observation benefit, total energy consumption, and task completion time of the satellite planning scheme in the embodiments of the present invention can be obtained.

[0161] S102. According to the judgment matrix, obtain the weight of each of the objectives.

[0162] After obtaining the above judgment matrix A, the weights of each objective can be calculated. Since the embodiments of the present invention consider the three objectives of the total observation benefit, total energy consumption, and task completion time of the satellite planning scheme, the weights of these three objectives are respectively denoted as w1, w2, and w3. Taking w1 as an example, its calculation method is shown in Equation (13):

[0163]

[0164] S20. For each of the bid proposals, normalize the values of its respective objectives in a vector normalization manner to obtain a normalized decision matrix.

[0165]

[0166] Among them, g represents the g-th bidding plan in this bidding process, assuming there are f bidding plans in total; k represents the k-th objective, and there are 3 objectives in total; c gk represents each objective corresponding to the g-th bidding plan in matrix C g , which are respectively the total observation benefit of the satellite planning plan, the total energy consumed, and the task completion time.

[0167] S30. According to the specified decision matrix, use the TOPSIS evaluation method to obtain the final index values of each of the said bidding plans; including:

[0168] S301. According to the specified decision matrix [z gk f×3 , obtain the ideal solution and the negative ideal solution

[0169]

[0170]

[0171] It should be noted that in the embodiment of the present invention, the total observation benefit objective of the satellite planning plan is a benefit-type objective, and the total energy consumption objective and the task completion time objective are both cost-type objectives.

[0172] S302. Calculate the weighted distances and between each of the said bidding plans and the ideal solution and

[0173]

[0174]

[0175] where w k represents the weight value of the K-th objective;

[0176] S303. According to the weighted distances and obtain the comprehensive evaluation indicators of each plan;

[0177]

[0178] where CE g represents the comprehensive evaluation indicator of the g-th bidding plan.

[0179] S40. Take the bidding plan with the highest final index value as the winning bid plan.

[0180] ​S5. The master satellite uses the multi-attribute bid evaluation strategy again to evaluate bids. For the current unassigned task sequence after the first update, it selects the winning bid from each of the bid proposals, updates the planning scheme of the second-winning slave satellite, and updates the currently unassigned task sequence for the second time. The multi-attribute bid evaluation strategy at least includes the total observation benefit of the satellite planning tasks.

[0181] S6. If all tasks have been arranged for observation, or there are no bid proposals, or the total observation benefit reaches the same number of loop termination times continuously, the planning ends and the globally optimal satellite task planning scheme is output; otherwise, let q = q + 1 and go to S2.

[0182] In the second aspect, as Figure 3 shown, the embodiment of the present invention provides a satellite task planning system based on a concurrent mechanism contract net, including several slave satellites and a master satellite for decision-making. The system includes:

[0183] A task acquisition module for executing S1. The master satellite acquires the task sequence to be observed, enters the loop bidding process, sets q = 1, and sets the loop termination times.

[0184] A bidding start module for executing S2. The master satellite starts the q-th bidding, acquires the currently unassigned task sequence. If the currently unassigned task sequence is empty, go to the scheme confirmation module to execute S6; otherwise, go to the scheme generation module S3.

[0185] A scheme generation module for executing S3. The master satellite broadcasts the currently unassigned task sequence to each of the slave satellites; according to the preset slave satellite mathematical model, each of the slave satellites generates the corresponding current bid proposal by using the greedy rule with the maximum benefit for the currently unassigned task sequence. If all the bid proposals are empty, go to the scheme confirmation module to execute S6; otherwise, go to the first scheme evaluation module to execute S4.

[0186] A first scheme evaluation module for executing S4. According to the preset master satellite mathematical model, the master satellite uses the multi-attribute bid evaluation strategy to evaluate bids, selects the winning bid from each of the bid proposals, updates the planning scheme of the first-winning slave satellite, and updates the currently unassigned task sequence for the first time. If the currently unassigned task sequence after this update is empty, go to the scheme confirmation module to execute S6; otherwise, go to the second scheme evaluation module to execute S5.

[0187] The second scheme bid evaluation module is used to execute S5. The master satellite uses the multi-attribute bid evaluation strategy again for bid evaluation. For the current unarranged task sequence after the first update, select the winning bid scheme from each of the bid schemes, update the planning scheme of the second-winning slave satellite, and update the current unarranged task sequence for the second time. The multi-attribute bid evaluation strategy at least includes the total observation benefit of the satellite planning tasks.

[0188] The scheme confirmation module is used to execute S6. If all tasks have been arranged for observation, or there is no bid scheme, or the total observation benefit reaches the same number of loop termination times continuously, the planning ends, and the globally optimal satellite task planning scheme is output; otherwise, let q = q + 1, and go to the tendering start module to execute S2.

[0189] In a third aspect, an embodiment of the present invention provides a storage medium that stores a computer program for satellite task planning based on the concurrent mechanism of the contract net. Among them, the computer program enables the computer to execute the satellite task planning method as described above.

[0190] In a fourth aspect, an embodiment of the present invention provides an electronic device, including:

[0191] One or more processors;

[0192] A memory; and

[0193] One or more programs, where the one or more programs are stored in the memory and are configured to be executed by the one or more processors. The programs include those for executing the satellite task planning method as described above.

[0194] In summary, compared with the prior art, the following beneficial effects are achieved:

[0195] The concurrent mechanism proposed in the embodiment of the present invention includes full-task tendering and a secondary bid evaluation strategy. By introducing the concurrent mechanism, the communication volume generated by the contract net protocol during the tendering process is reduced, and the problem of large communication volume in traditional contract net negotiation is overcome; in addition, the bid evaluation criteria for using the contract net to solve the multi-satellite and multi-task allocation problem are improved through the multi-attribute bid evaluation strategy.

[0196] It should be noted that in this text, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "comprising an..." does not exclude the existence of additional identical elements in the process, method, article or device comprising the said element.

[0197] The above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A satellite mission planning method for a contract net based on a concurrency mechanism, characterized in that, Including several slave satellites and a master satellite for decision-making, the method includes: S1. The master satellite obtains the sequence of tasks to be observed, enters a cyclic bidding process, sets q = 1, and sets the number of times to terminate the cycle. S2. The master satellite starts the q-th bidding, obtains the sequence of tasks that have not been arranged currently. If the sequence of tasks that have not been arranged currently is empty, go to S6; otherwise, go to S3. S3. The master satellite broadcasts the sequence of tasks that have not been arranged currently to each slave satellite; according to the preset mathematical model of the slave satellite, each slave satellite generates a corresponding current bidding plan for this sequence of tasks that have not been arranged currently by using the greedy rule with the maximum benefit. If all the bidding plans are empty, go to S6; otherwise, go to S4. S4. According to the preset mathematical model of the master satellite, the master satellite uses a multi-attribute bid evaluation strategy to evaluate the bids, selects the winning bid from each bidding plan, and updates the planning plan of the first winning slave satellite, and updates the sequence of tasks that have not been arranged currently for the first time. If the sequence of tasks that have not been arranged currently after this update is empty, go to S6; otherwise, go to S5. S5. The master satellite uses the multi-attribute bid evaluation strategy again to evaluate the bids. For the sequence of tasks that have not been arranged currently after the first update, selects the winning bid from each bidding plan, and updates the planning plan of the second winning slave satellite, and updates the sequence of tasks that have not been arranged currently for the second time; the multi-attribute bid evaluation strategy at least includes the total observation benefit of the satellite planning tasks. S6. If all tasks have been arranged for observation, or there is no bidding plan, or the total observation benefit reaches the same number of times of the cycle termination continuously, the planning ends, and the globally optimal satellite task planning plan is output; otherwise, set q = q + 1, and go to S2. In S4, when the master satellite uses a multi-attribute bid evaluation strategy to evaluate the bids and selects the winning bid from each bidding plan, it specifically includes: S10. Use the AHP subjective weighting method to determine the weight of each preset goal. The goals include the total observation benefit of the satellite planning plan, the total energy consumed, and the task completion time. S20. For each bidding plan, normalize the values of its respective goals in a vector normalization manner to obtain a normalized decision matrix. S30. According to the normalized decision matrix, use the TOPSIS bid evaluation method to obtain the final index value of each bidding plan. S40. Take the bidding plan with the highest final index value as the winning bid.

2. The satellite mission planning method according to claim 1, wherein The mathematical model of the slave satellite in S3 includes an objective function for maximizing the benefit: And the first constraint condition: OTS ij +dur j =OTE ij (4) e ij = e i dur j , i ∈ {1, …, m} (6) Among them, formula (2) represents the uniqueness constraint, that is, a task can be observed at most once; formula (3) represents that the time window requirement must be met when the task is observed; formula (4) represents the relationship between the actual observation end time of the task and the task observation duration; formulas (5 - 6) represent the energy constraint and the energy consumption calculation method in satellite planning; formula (7) represents the storage constraint. The main satellite S′, the set of slave satellites S = {S1, S2, …, S i , …, S m}, with a total of m slave satellites; T = {t1, t2, …, t j , …, t n} represents the sequence of tasks to be observed, with a total of n tasks; t j = <P j , dur j , dt j , that is, the attributes of the task are represented by a triple, where P j represents the observed benefit of task t j , dur j represents the observed duration of task t j , dt j represents the task observation deadline of task t j ; Represents the set of visible time windows, representing task t j Among the visible time window sets from satellite S i The set of visible time windows on it, NTM ij Represents the number of time windows That is, the visible time window is represented by a binary tuple, representing task t j in the k-th visible time window from satellite S i and and respectively represent the start time and end time of the visible time window; OTW ij = <OTS ij , OTE ij >, that is, the observation time window is represented by a binary tuple. OTW ij represents the actual observation time window of task t j from satellite S i and OTS ij and OTE ij represent the start time and end time of this observation time window respectively; W p and W r represent the weights of the observed benefit and the perturbation, respectively; e i represents the unit energy consumption of the observation mission from satellite S i ; e ij Indicates the energy consumed i from the satellite S j for the observation mission t; E i represents the maximum energy limit from satellite S i ; c ij Indicates the storage consumed for i observation mission t j from satellite S; C i Indicates the maximum storage limit from satellite S i ; Indicates task t j whether to observe during the k-th visible time window from satellite S i If so, the value is 1; otherwise, the value is 0.

3. The satellite mission planning method according to claim 2, wherein The mathematical model of the master satellite in S5 includes Objective function one for maximizing the total observation benefit: Objective function two for minimizing the total energy consumed: Objective function three for minimizing the task completion time of the satellite planning scheme: And the second constraint condition: ET iq ≤PTE, where i ∈ {1, …, m} and q ∈ {1, 2, …} (12) Among them, formula (11) represents the constraint between the number of tasks during the bidding process; formula (12) represents the planning time deadline constraint; ET iq Indicates the observation end time of the solution generated for the mission sequence T from satellite S in the q-th tendering and bidding, and i for the mission sequence T q and where lst i represents the observation end time of the last mission on satellite S; PTE represents the mission planning deadline. i ​ 4. The satellite mission planning method according to claim 1, wherein In S10, the AHP subjective weighting method is used to determine the weight of each preset objective, specifically including: S101. Construct a judgment matrix according to each of the objectives and a preset objective quantification value standard; S102. Obtain the weight of each of the objectives according to the judgment matrix.

5. The satellite mission planning method according to claim 1, wherein S30 specifically includes: S301. Obtain the ideal solution and the negative ideal solution corresponding to each target according to the specified decision matrix [z gk f×3 ​​​ Among them, g represents the g-th bidding plan in this bidding, and there are f bidding plans in total; k represents the k-th objective, and there are 3 objectives in total; c gk represents each objective corresponding to the g-th bidding plan in matrix C g respectively the total observation revenue, the total energy consumed, and the mission completion time of the satellite planning plan; S302. Calculate the weighted distances between each of the tendering solutions and the ideal solution and the negative ideal solution respectively and Among them, w k represents the weight value of the Kth target; S303. Obtain the comprehensive bid evaluation index of each plan according to the weighted distance and ; Among them, CE g represents the comprehensive bid evaluation index of the g-th bidding plan.

6. The satellite mission planning method according to any one of claims 1 to 5, characterized in that, After the main satellite in S1 obtains the sequence of tasks to be observed, the sequence of tasks to be observed is re-sorted from largest to smallest according to the benefit; if there are multiple tasks with the same benefit, a secondary sorting is performed from smallest to largest according to the task observation duration, and then it enters the loop bidding process.

7. A satellite mission planning system for a contract network based on a concurrency mechanism, characterized in that, Including several slave satellites and a main satellite for decision-making. This system is used to execute the satellite task planning method as described in claim 1, including: A task acquisition module, used to execute S1, the main satellite obtains the sequence of tasks to be observed, enters the loop bidding process, sets q = 1, and sets the loop termination times; A bidding start module, used to execute S2, the main satellite starts the qth bidding, obtains the sequence of tasks that have not been arranged currently. If the sequence of tasks that have not been arranged currently is empty, transfer to the scheme confirmation module to execute S6; otherwise, transfer to the scheme generation module S3; A scheme generation module, used to execute S3, the main satellite broadcasts the sequence of tasks that have not been arranged currently to each of the slave satellites; according to the preset slave satellite mathematical model, each of the slave satellites generates a corresponding current bidding scheme for the sequence of tasks that have not been arranged currently by using the greedy rule with the maximum benefit. If all the bidding schemes are empty, transfer to the scheme confirmation module to execute S6; otherwise, transfer to the first scheme evaluation module to execute S4; The first scheme evaluation module, used to execute S4, according to the preset main satellite mathematical model, the main satellite uses a multi-attribute evaluation strategy for evaluation, selects the winning scheme from each of the bidding schemes, and updates the planning scheme of the first winning slave satellite, and updates the sequence of tasks that have not been arranged currently for the first time. If the sequence of tasks that have not been arranged currently after this update is empty, transfer to the scheme confirmation module to execute S6; otherwise, transfer to the second scheme evaluation module to execute S5; The second scheme evaluation module, used to execute S5, the main satellite uses the multi-attribute evaluation strategy for evaluation again, selects the winning scheme from each of the bidding schemes for the sequence of tasks that have not been arranged currently after the first update, and updates the planning scheme of the second winning slave satellite, and updates the sequence of tasks that have not been arranged currently for the second time; the multi-attribute evaluation strategy at least includes the total observation benefit of the satellite planning tasks; The scheme confirmation module, used to execute S6, if all tasks have been arranged for observation, or there is no bidding scheme, or the total observation benefit reaches the same loop termination times continuously, then the planning ends, and the globally optimal satellite task planning scheme is output; otherwise, set q = q + 1, and transfer to the bidding start module to execute S2.

8. A storage medium, characterized in that, It stores a computer program for satellite mission planning based on the contract net of the concurrency mechanism, wherein the computer program causes a computer to execute the satellite mission planning method according to any one of claims 1 to 6.

9. An electronic device, characterized in that, Comprising: One or more processors; A memory; And one or more programs, wherein the one or more programs are stored in the memory and configured to be executed by the one or more processors, and the programs include those for executing the satellite mission planning method according to any one of claims 1 to 6.

Images