CN115081884B

CN115081884B - Distributed on-board online many-to-many task planning method

Info

Publication number: CN115081884B
Application number: CN202210723636.3A
Authority: CN
Inventors: 马广富; 徐杭; 郭延宁; 吕跃勇; 沈锐臻; 陈雪松
Original assignee: Harbin Institute of Technology
Current assignee: Harbin Institute of Technology
Priority date: 2022-06-23
Filing date: 2022-06-23
Publication date: 2023-05-02
Anticipated expiration: 2042-06-23
Also published as: CN115081884A

Abstract

A distributed on-line multi-to-multi-task planning method belongs to the field of task planning of service spacecraft. The method solves the problems of slow speed of generating the scheme and long response time of dynamic scheme adjustment in the existing method. The scheme of the invention is as follows: step one, collecting task demands and calculation time demands of on-orbit service, and calculating maximum iteration round parameters of inner auction; step two, based on the calculated maximum iteration round parameter of the inner auction, the tasks of the on-orbit service are distributed under the fuel constraint of each service spacecraft, and all task execution sequence lists and orbit transfer process duration information are obtained; and thirdly, correcting the speed pulse of each stage in the track change process by considering the influence of the power in the track change process, and obtaining a final planning scheme according to the task execution sequence list, the track change process duration information and the corrected speed pulse of each stage. The method can be applied to task planning of the service spacecraft.

Description

Distributed on-board online many-to-many task planning method

Technical Field

The invention belongs to the field of task planning of service spacecrafts, and particularly relates to a distributed on-board online multi-to-multi-task planning method.

Background

With the increasing number of human aerospace activities, the number of in-orbit satellites and spacecraft has been very large, and by the time of 2021, 12 months and 31 days, 4852 satellites are available worldwide, and the total number of Low Earth Orbit (LEO) satellites is 4078, which is the majority. As a high value product, when satellite is lost and fuel is insufficient, the function maintenance of the satellite by using the on-orbit service mode has higher economic benefit compared with retransmission. Meanwhile, space satellites, particularly LEO satellites, face a great threat due to the existence of space fragments, at least 26000 fragments larger than 10cm and even larger, and more than 1 hundred million fragments smaller than 1 mm. According to the statistics, the mass of fragments in space exceeds 8000 tons and is distributed on each track. Studies have shown that collisions between fragments will create a cascading effect when LEO rail object density reaches a certain level, causing large bursts once the critical point is accumulated, and thus cleaning up space fragments is not slow.

Whether space on-orbit maintenance and upgrading are carried out facing to satellites, fuel oil filling is carried out, or space debris is cleaned, a spacecraft is used for active orbit transfer and then service is applied. Aiming at a plurality of on-orbit tasks, the on-orbit tasks are required to be distributed to a plurality of spacecrafts to be carried out simultaneously, the distribution relation, the service time sequence, the orbit changing process, the transfer window, the transfer time length and the speed pulse of each orbit changing time of the plurality of spacecrafts are required to be determined under the consideration of perturbation, and then the scheme is sent to the plurality of spacecrafts to be executed.

The existing method generally gathers requirements on a ground station, and after a scheme is formed, instructions are uploaded to a service spacecraft to enable the spacecraft to start maneuvering and orbit transfer, and service is prepared. However, the scheme generation at the ground station is often not as fast as the on-board scheme generation due to the reasons of communication time delay, measurement and control time window waiting and the like; and once the space state is changed, such as secondary collision of space fragments, the real-time dynamic scheme adjustment response time is long. Therefore, an intuitive idea is to perform scheme generation on the satellite, but due to weight and cost limitation of the service spacecraft, on-board planning is more sensitive to the calculation amount of the generation strategy, and if the calculation process on the ground station is directly transplanted to on-board calculation, the requirement of rapid scheme generation cannot be met.

Disclosure of Invention

The invention aims to solve the problems of slow speed of the existing method generation scheme and long response time of dynamic scheme adjustment, and provides a distributed on-board online many-to-many task planning method.

The technical scheme adopted by the invention for solving the technical problems is as follows:

a distributed on-line multi-to-multi-task planning method specifically comprises the following steps:

step one, collecting task requirements and calculation time requirements of on-orbit service, and calculating an inner layer auction maximum iteration round parameter pnum according to the task requirements and the calculation time requirements of the on-orbit service;

step two, based on the calculated maximum iteration round parameter pnum of the inner layer auction, distributing tasks of on-orbit service under the fuel constraint of each service spacecraft to obtain an execution sequence list of all tasks and orbit transfer process duration information;

and thirdly, correcting the speed pulse of each stage in the track change process by considering the influence of the power in the track change process, and obtaining a final planning scheme according to the task execution sequence list, the track change process duration information and the corrected speed pulse of each stage.

The beneficial effects of the invention are as follows:

under the condition that the on-board computing capacity of the spacecraft is tension, and meanwhile, a planning scheme is required to be formed rapidly in face of an emergency, a two-layer distributed auction algorithm is adopted, and the LEO orbit perturbation is considered to rapidly generate the planning scheme, so that the method has the characteristics of rapid response time, low operation consumption and self-adaptive and adjustable optimization time of dynamic scheme adjustment, and can enable multi-satellite on-orbit service autonomous planning subsequently. By generating the scheme on the satellite, the scheme can be generated quickly.

Drawings

FIG. 1 is a flow chart of a distributed on-board online many-to-many task planning method of the present invention;

FIG. 2 is a graph of the number of orbit calculations versus the number of service spacecraft;

FIG. 3 is a graph of the number of track-change calculations versus the number of tasks;

FIG. 4 is a diagram of a three-step derailment strategy derailment process;

FIG. 5 is a schematic diagram of a Homan transfer process;

FIG. 6 is a schematic diagram of an out-of-plane derailment process;

FIG. 7 is a schematic diagram of a phase modulation maneuver;

FIG. 8 is a schematic diagram of a Lambert transfer process;

FIG. 9 is a graph of the average change in inner layer velocity delta consumption;

FIG. 10 is a graph of the change in the outer layer velocity delta consumption optimum average.

Detailed Description

Detailed description of the inventionin the first embodiment, this embodiment will be described with reference to fig. 1. The method for planning the distributed on-line multi-to-multi-task on the satellite in the embodiment specifically comprises the following steps:

In this embodiment, on-board online many-to-many mission planning refers to a multi-mission, multi-service spacecraft.

The second embodiment is as follows: the first embodiment is different from the first embodiment in that the specific process of the first step is:

wherein pnum represents the maximum iteration round of the inner auction, T ₀ Representing the operation time required by 10 service spacecrafts and 10 tasks estimated in advance, wherein T represents the calculation time constraint on the satellite, namely the calculation time requirement; n represents the actual number of service spacecraft, and m represents the actual number of tasks, i.e. task demands.

Because the on-board calculation resources are limited, the number of iterative rounds which can be allowed to be calculated on the premise of meeting calculation time constraint needs to be estimated in advance, so that the iterative rounds can be continuously optimized in a subsequent algorithm. The spacecraft determines a task set and a spacecraft set of the auction through communication among the spacecraft, and the spacecraft which select to release the tasks perform calculation processes such as summarizing, distributing and the like. Because the main time in the task is concentrated in a large number of track change calculation processes, the calculated magnitude can be estimated approximately by taking the three steps of track change strategy calculation times as indexes. As shown in fig. 2 and 3, the number of iterations of the two layers is set to be 1, and the change relation between the number of times of three steps of orbit transfer strategy calculation and the number of service spacecrafts and the number of tasks in the algorithm calculation process is drawn.

In fig. 2, under the same task number, the number of service spacecrafts is considered to be 1-50, and the number of times of orbit transfer calculation after operation is finished is taken to draw a curve. In fig. 3, under the condition of the same number of service spacecrafts, taking the number of tasks of 1-100, and drawing a curve by taking the number of times of orbit transfer calculation after operation. From the above results, the calculated amount of the algorithm and the number of service spacecraft and the number of tasks are approximate to be linear. The operation time T of 10 spacecrafts and 10 tasks can be estimated in advance in actual use ₀ The approximate iteration inner layer round pnum satisfies:

because the actual calculation has deviation, the inner layer is mainly constrained by the design rounds, the outer layer considers the actual calculation time constraint, the time design advance is calculated according to the historical calculation time of each round of the outer layer, and the iterative calculation is stopped once the next round of calculation is estimated to exceed the time limit.

Other steps and parameters are the same as in the first embodiment.

And a third specific embodiment: the difference between this embodiment and the first or second embodiment is that the specific process of the second step is:

step two, initializing outer layer iteration rounds k=1, and randomly generating auction sequences of n service spacecrafts;

secondly, determining task priority ordering rules of each round in each round of auction iteration according to the inner layer auction iteration round parameters;

initializing an inner auction iteration round p=1, and sorting task target priorities according to the task priority sorting rule determined in the second step to generate an auction task list of each round;

the number of tasks of each iteration is smaller than or equal to the number of service spacecrafts, and each iteration service spacecrafts select at most one task or do not select tasks;

step two, according to the auction sequence determined in the step two, n service spacecrafts sequentially select the maximum surplus task from the auction task list of the first round;

if the fuel of the service spacecraft cannot meet the price spent by the service spacecraft (namely, the fuel consumption required by the service spacecraft is larger than the fuel capable of being provided by the service spacecraft) when the maximum surplus task is selected to be executed, the service spacecraft does not select the task; otherwise, the service spacecraft selects the maximum surplus task and updates the prices of the selected task relative to all service spacecraft (in the current round, the subsequent spacecraft performs task selection based on the updated prices), and judges whether the selected task has been selected by other service spacecraft;

if the selected task has been selected by other service spacecraft, the original allocation service spacecraft of the selected task reselects the maximum surplus task, and judges whether the fuel meets the condition or not and updates the price (if the re-selected task of the original allocation service spacecraft is still selected by other service spacecraft at the back, the original allocation service spacecraft reselects the maximum surplus task again until no repeated selected event of the task occurs or the service spacecraft cannot divide the task under the constraint premise, and if the current round has the residual task, the next service spacecraft continues to select the task); if the selected task is not selected by other service spacecrafts, the selected task does not need to be processed;

step five, after the number of rounds is increased by 1, returning to the process of executing the step two, until all tasks are distributed or no service spacecraft in the current round takes the tasks; executing the second step;

step six, making the inner layer auction iteration round p=p+1, returning to the step two, repeatedly executing the step two, three and the step two, five until the inner layer auction iteration round reaches the maximum iteration round of the inner layer auction, and executing the step two, seven;

step seven, making the outer layer iteration round k=k+1, returning to the step two, and repeatedly executing the processes of the step two to the step two, until the operation time requirement is met, and executing the step two, eight;

and step two, selecting an optimal task allocation scheme according to the surplus scores of successful tasks to obtain a time sequence task list for executing all tasks and time length information of each task track-changing process (Homan track-changing, out-of-plane track-changing and phase modulation maneuver).

And obtaining a task allocation scheme after each round of inner auction iteration is executed, wherein each round of inner auction iteration corresponds to a plurality of rounds, and if a certain service spacecraft selects a task in different rounds, the service spacecraft executes according to the sequence of the time for selecting the task. And calculating the surplus scores of the successful tasks corresponding to each task allocation scheme.

Other steps and parameters are the same as in the first or second embodiment.

The specific embodiment IV is as follows: the difference between the present embodiment and the first to third embodiments is that the specific process of the second step is:

if the inner auction iteration round is the first round, the task priority ordering rule in each round of the first round is: sequencing tasks according to the order of the optimal surplus from big to small; if the inner auction iteration round is not the first round, randomly generating the ordering of the tasks in each round;

the number of tasks in each turn is less than or equal to the number n of service spacecraft.

Other steps and parameters are the same as in one to three embodiments.

Fifth embodiment: the difference between this embodiment and one to four embodiments is that the calculation rule of the optimal surplus is: for any task, calculating surplus obtained by each service spacecraft for serving the task, and taking the maximum value of the surplus as the optimal surplus of the task;

the surplus obtained by the service task j of the service spacecraft i is as follows:

G _ij ＝E _ij -C _ij

wherein ,E_ij Representing the benefit obtained by service task j of service spacecraft i, C _ij Representing the price spent servicing task j by service spacecraft i, G _ij Representing surplus obtained by a service task j of the service spacecraft i;

the price is defined as: when a spacecraft service task is serviced, fuel consumption (speed increment consumption) required by orbital transfer is carried out by adopting a three-step orbital transfer strategy, and the price is expressed as:

C _ij ＝Δv _hm +Δv _ym +Δv _tx

wherein ,Δv_hm Representing the incremental consumption of the speed of the Homan change of track, deltav _ym Indicating the speed increment consumption of the out-of-plane track change, deltav _tx Indicating the incremental consumption of speed of the phase modulator maneuver.

In particular, only the disomic dynamics, i.e. the action of the earth's attraction, is considered here for accelerating the calculation, and the uptake dynamics is not considered.

Other steps and parameters are the same as in one to four embodiments.

Specific embodiment six: the difference between the embodiment and one to fifth of the specific embodiments is that the three-step orbit transfer strategy includes three stages of huffman orbit transfer, out-of-plane orbit transfer and phase modulation maneuver, the service spacecraft sequentially performs the huffman orbit transfer, out-of-plane orbit transfer and phase modulation maneuver, and the specific orbit transfer process is as follows:

the initial position of the service spacecraft is marked as a point A, and the service spacecraft realizes the same orbit height with the target at a point B through a Homan orbit change process at the initial position point A; when the service spacecraft continues to travel to the intersection point C of the orbit surface of the service spacecraft and the target orbit surface, service spacecraft deployment different-surface rail changing; after the different-surface orbit is changed, the phase angle difference adjustment between the service spacecraft and the target is realized through the phase modulation maneuvering process, and the service spacecraft meets the target.

The embodiment designs a three-step orbit transfer strategy, and the Homan orbit transfer, the out-of-plane orbit transfer and the phase modulation are sequentially carried out, so that the service spacecraft can meet any orbit targets on the premise that all task target orbits are approximately circular orbits.

As shown in fig. 4, the spacecraft is initially at a, and the same orbit height as the target is realized at B through a huffman orbit transfer process (1); at the intermediate track (2) running to the track intersection point C, carrying out different-surface track changing to realize the same track surface as the target; finally, phase angle difference adjustment with the target is realized through a phase modulation maneuvering process (3), and intersection with the target is realized.

In the phase modulation maneuver (3), particularly for LEO orbiting satellites, it is necessary to determine whether the orbit maneuver will strike the earth, so the orbit maneuver needs to satisfy:

r _min ＞r _e

if the condition cannot be satisfied, the task is considered to exceed the service capability.

The specific process of the Homan orbit transfer is as follows:

the huoman transfer can realize the change of track height, and aims at two coplanar circular track transfers with a common focus, the transfer is the double-pulse maneuvering transfer with the most energy saving, and has smaller burnup at the expense of track change flexibility relative to Lambert track change. As shown in fig. 5, the huffman transfer elliptical orbit is tangent to both the round orbit before and after transfer on the archwire.

The huoman transfer process can be transferred from a low orbit to a high orbit, as shown in a process (1) in fig. 5, the spacecraft needs to provide a speed increment in the flight direction at a, an increasing speed value is switched to a transfer orbit, after a half transfer orbit period is operated, the speed increment is continuously provided in the flight direction at B, and the increasing speed value is switched to a target orbit; the spacecraft may also be transferred from the high rail to the low rail, as shown in the process (2) in fig. 5, where the spacecraft needs to provide a speed increment in the flight direction at B, the reduced speed value is switched to the transfer rail, and after half a transfer rail period of operation, the speed increment in the flight direction is continuously provided at a, and the reduced speed value is switched to the target rail. The phase angle of the spacecraft in the transfer process is increased by 180 degrees, and the transfer time is half of the Homan transfer orbit period.

The incremental speed consumption of the transfer process is:

wherein G represents a constant of gravitational force, M _e Representing earth mass, r _A and r_B Representing the radius of the circular orbit before and after the huffman transfer. The orbital sharing of the service spacecraft and the target can be realized through the Homan orbital transfer, so that the intersection point exists between the spacecraft orbit plane and the target orbit plane.

Different-surface rail

The out-of-plane orbit change is to change the current orbit speed in a speed pulse mode, so that the orbit is switched into a target orbit, and the service spacecraft and the target run in the same orbit plane, as shown in fig. 6:

from the knowledge of the sphere triangle correlation, it can be seen that:

cosθ＝cosβ ₁ cosβ ₂ +sinβ ₁ sinβ ₂ cosΔΩ

ΔΩ＝Ω ₂ -Ω ₁

the speed increment consumed by the out-of-plane track change is calculated as follows:

wherein r represents the radius of the intersecting circular orbit, Ω ₁ and Ω₂ The ascent and intersection points of two track surfaces are represented by the right ascent and intersection points, beta ₁ and β₂ The track inclination of the two track surfaces is shown, and θ represents the angle between the two track surfaces.

Phase modulation machine

The phase modulation maneuver can be considered as a double pulse huffman transfer, just leaving and returning to the same orbit, and the transfer orbit is an entire elliptical period.

As shown in fig. 7, the position of the spacecraft on the circular orbit can be changed by phase modulation maneuver, thereby reducing the phase angle difference between the spacecraft and the target. When the target is advanced in the spacecraft, for example, the spacecraft is initially at the position a and the target is at the position B, the target can be caught up with the phase modulation orbit (1) in fig. 7, and the spacecraft needs to firstly decelerate at the position a, enter the phase modulation orbit, then run an elliptical period to return to the position a and then accelerate again, and return to the original orbit. When the target is behind the spacecraft, for example, the spacecraft is initially at the position a and the target is at the position C, the phase modulation orbit (2) in fig. 7 can be adopted to wait for the target, and the spacecraft needs to accelerate at the position a, enter the phase modulation orbit, then run an elliptical period to return to the position a and then decelerate, and return to the original orbit.

During phase modulation and orbit transfer, the service spacecraft applies two velocity pulses at the initial position, and the consumed velocity increment is:

wherein r represents the radius of the original running track, a represents the semi-major axis of the transfer elliptical track, and n _t and n_s Respectively represent the number of turns of the orbit of the target and the service spacecraft in the phase modulation maneuver, the time rapidity is considered in the invention, the single-turn phase modulation maneuver is adopted,

representing the phase difference between the service spacecraft and the target before orbit transfer, taking the target flight direction as the positive direction, and taking the angle from the target to the service spacecraft as the phase difference, < + >>

Other steps and parameters are the same as in one of the first to fifth embodiments.

Seventh embodiment: this embodiment differs from one to six embodiments in that the updating the price of the selected task with respect to all service spacecraft is performed by:

if a certain service spacecraft i ₀ Selecting task j in the current round ₀ Then the other service spacecraft i' service tasks j ₀ The price to be spent is updated as:

wherein i' noteqi ₀ ，

Is the updated service task j of the service spacecraft i ₀ Price of spending->

Is the service task j of the service spacecraft i' before updating ₀ Price spent, j' is any task in the current round,/>

Is the surplus value corresponding to the task j 'with the largest surplus in the current round of service of the service spacecraft i', and is->

The method is that the service spacecraft i 'serves the surplus value corresponding to the surplus second largest task j' in the current round, and alpha is the rising price parameter;

if there is no surplus second largest task j', then

/>

Other steps and parameters are the same as in one of the first to sixth embodiments.

Eighth embodiment: the difference between this embodiment and one of the first to seventh embodiments is that the calculation method of the surplus score of the successful task is as follows:

for a certain task allocation scheme, calculating a successful task surplus score P corresponding to the task allocation scheme:

wherein ,

average value, n, representing surplus obtained by successful tasks in such a task allocation scheme _f Representing the number of failed tasks, p _t Representing an adjustment score for each failed task.

Other steps and parameters are the same as those of one of the first to seventh embodiments.

Detailed description nine: the difference between the embodiment and one to eighth embodiments is that after the optimal scheme is formed by the two-layer distributed auction algorithm, the spacecraft generating the scheme issues the scheme to each service spacecraft which is specifically executed, and each service spacecraft carries out orbital transfer process correction by considering the ingestion power for the task which is distributed to the service spacecraft. Under the influence of the perturbation force, the orbit surface of the satellite is also changed, and the actual orbit motion process is recursively calculated according to the following Dragon-Kutta method and the perturbation force influence formula:

wherein r is an object position vector, a _TBP For trisomy perturbation acceleration, a _NSP Is the global aspheric perturbation acceleration, a _ADP A is atmospheric resistance perturbation acceleration _F Acceleration provided for the external thrust of the fuel.

The speed pulse correction method in the Huffman orbit transfer process is as follows:

the spacecraft obtains a first pulse of the two-dimensional dynamics calculation, orbit information of the spacecraft and a target is recursively calculated through a formula (1), the recursion time length is obtained, the orbit change time length obtained by the two-dimensional dynamics is increased by a time length delta T (which can be selected according to actual conditions), the recursion time length is divided according to step sizes, the difference between the orbit radius of the spacecraft and the orbit radius of the target at each step size interval is analyzed, and the minimum difference is obtained.

When the two sides move to the minimum difference, correcting the initial first pulse according to the initial track height and the nearest distance height of the two sides:

v′ _hm1 ＝v _hm1 (1+γd _min )

in the formula ,v_hm1 The velocity vector of the spacecraft after being sprayed by the first Howman orbital transfer velocity pulse is consistent with the original orbital velocity direction, v _h ′ _m1 The velocity vector after the first Howman orbital transfer velocity pulse ejection of the corrected spacecraft is represented, gamma represents a set coefficient, d _min Representing the radius difference, two cases:

a. target track height at minimum difference, d _min Unchanged;

b. target track is low at minimum difference, d _min Taking the opposite number;

and after the first correction, starting the next track recursion, and when the minimum difference value is smaller than the set threshold value, considering that iteration is ended, wherein the corresponding track change duration is the Huffman track change correction track change duration. And after the task orbit height is reached, acquiring the orbit speed according to six real orbits of the recursion final state spacecraft, and comparing the orbit speed with the recursion final state speed to acquire a second speed pulse.

The speed pulse correction mode in the out-of-plane track change process is as follows:

and (3) recursing orbit information of the spacecraft and the target through a formula (1), updating orbit normal vectors and orbit intersection points of the spacecraft and the target in real time, and taking a dot product minimum value of a connecting line vector of the two orbit intersection points and a spacecraft position vector, wherein the connecting line of the corresponding spacecraft position vector and the intersection point coincides, so that a different-plane orbit change process time window is obtained.

According to six real orbits of the recursion last-state spacecraft, the orbit normal vector is changed to be consistent with the target, the orbit speed is obtained, and the orbit speed is compared with the recursion last-state speed, so that a third out-of-plane orbit speed pulse is obtained.

The speed pulse correction mode of the phase modulation maneuver is as follows:

in practice, after the recursion of the first two steps is considered, it is difficult to ensure that the target orbits of the spacecraft are highly consistent, and in order to reduce errors, the speed pulse correction process of phase modulation maneuver is realized through Lambert orbit transfer, where phase modulation maneuver can be regarded as a special process of Lambert orbit transfer.

After the first speed pulse is made by Lambert, comparing the deviation between the true position and the expected position after the track-changing time length is recursively calculated, and negatively feeding back the deviation to the target position r calculated by Lambert iteration _C And (3) until the target position and the expected position are smaller than a threshold value, obtaining the initial speed vector and the final speed vector of the Lambert orbit transfer process, comparing the initial speed vector of the Lambert orbit transfer process with the speed vector at the end of the different-surface orbit transfer process, and comparing the speed vector at the end of the Lambert orbit transfer process with the target orbit stable speed vector of the phase modulation maneuvering process, thus obtaining the two speed pulses of the Lambert orbit transfer process.

Meanwhile, the error generated in the iteration process is considered, the track-changing time length is changed, the condition of minimum fuel consumption is optimized and found nearby the track-changing time length, and the track-changing time length is used as the final correction speed pulse and the track-changing time length.

Lambert orbit determination

The Lambert orbit transfer process can realize orbit transfer between any two points in space, as shown in fig. 8.

The Lambert orbit transfer process mathematical equation can be expressed as:

(v _A ,v _C )＝f(r _A ,r _C ,T _trans )

in the method, the vectors of the two points before and after transfer are respectively r _A And r _C Speed v _A And v _C The movement duration of the track transfer process is T _trans . The mathematical equation of the Lambert orbit transfer process can obtain an analytic solution by a numerical iteration method, and the space object position vector r and the orbit velocity v ₀ With the six main points of the trackThe relation of the element phi is as follows:

(v ₀ ,r)＝f(φ)

the Lambert transfer process speed delta consumption can thus be expressed as:

Δv _lm ＝|v _A -v _A0 |+|v _C -v _C0 |

the main parameter definitions in the present invention are shown in table 1:

TABLE 1

/>

Relevant parameters and simulation results of the invention

1. Related parameters

In the test, 10 groups of service spacecrafts are set, and parameters of the service spacecrafts are shown in table 2:

table 2 orbit parameters for each service spacecraft

Task goal settings 20 groups, track parameter settings as shown in table 3:

TABLE 3 target track parameters for each task

/>

The other test parameters were set as shown in table 4:

table 4 algorithm test parameters

2. Simulation results

The final scheme is obtained by iterative operation, wherein the scheme before perturbation effect correction is shown in tables 5 and 6:

TABLE 5

/>

TABLE 6

The modified scheme of perturbation effect is shown in tables 7 and 8:

TABLE 7

/>

TABLE 8

For a certain group of spacecraft auction orders, the relationship between the average consumption value of each speed increment and the iteration times in the inner layer iterative optimization process is shown in fig. 9.

It can be seen from fig. 9 that after multiple rounds of random optimization, the primary selection effect can be significantly improved, and the possibility of generating a smaller fuel consumption average scheme is provided. And on the premise of time permission, searching for the solution results for multiple times is helpful to optimize the solution results.

Meanwhile, aiming at a certain outer layer iterative optimization process, the relation between the consumption average value of each round of optimal speed increment and the iterative times is drawn, as shown in fig. 10:

and as the outer layer iteration turns are increased, the outer layer speed increment consumption average value is optimally and gradually reduced, namely the fuel consumption of the generating scheme is gradually reduced. However, the initial iteration has a better result from the change of the indication, and when the time is sufficient, the scheme can be more energy-saving by multiple iterations, but if the time is tense, only one iteration can be performed, and the generation scheme is relatively ideal.

The above examples of the present invention are only for describing the calculation model and calculation flow of the present invention in detail, and are not limiting of the embodiments of the present invention. Other variations and modifications of the above description will be apparent to those of ordinary skill in the art, and it is not intended to be exhaustive of all embodiments, all of which are within the scope of the invention.

Claims

1. The on-line multi-to-multi-task planning method for the distributed satellite is characterized by comprising the following steps of:

step two, based on the calculated maximum iteration round parameter pnum of the inner layer auction, the tasks of the on-orbit service are distributed under the fuel constraint of each service spacecraft, and all task execution sequence lists and the orbit transfer process duration information are obtained, wherein the specific process is as follows:

step two, determining task priority ordering rules of each round in each round of auction iteration according to the inner layer auction iteration round parameters, wherein the specific process is as follows:

the number of tasks in each round is less than or equal to the number n of service spacecraft;

if the fuel of the service spacecraft cannot meet the price spent by the service spacecraft when the maximum surplus task is selected to be executed, the service spacecraft does not select the task; otherwise, the service spacecraft selects the maximum surplus task, updates the prices of the selected task relative to all service spacecraft, and judges whether the selected task is already selected by other service spacecraft;

if the selected task is already selected by other service spacecrafts, the original allocation service spacecrafts of the selected task reselect the maximum surplus task, and judge whether the fuel meets the condition and update the price; if the selected task is not selected by other service spacecrafts, the selected task does not need to be processed;

step two, selecting an optimal task allocation scheme according to the surplus scores of successful tasks to obtain a time sequence task list for executing all tasks and time length information of the track-changing process of each task;

2. The method for online multi-to-multi-task planning on a distributed satellite according to claim 1, wherein the specific process of the first step is as follows:

wherein pnum represents the maximum iteration round of the inner auction, T ₀ Representing the operation time required by 10 service spacecrafts and 10 tasks estimated in advance, wherein T represents the calculation time constraint on the satellite, n represents the actual number of the service spacecrafts, and m represents the actual number of the tasks.

3. The method for online multi-to-multi-task planning on a distributed satellite according to claim 2, wherein the calculation rule of the optimal surplus is: for any task, calculating surplus obtained by each service spacecraft for serving the task, and taking the maximum value of the surplus as the optimal surplus of the task;

G _ij ＝E _ij -C _ij

wherein ,E_ij Representing the revenue obtained by servicing the spacecraft i service mission j,C _ij representing the price spent servicing task j by service spacecraft i, G _ij Representing surplus obtained by a service task j of the service spacecraft i;

the price is defined as: when a spacecraft service task is serviced, adopting a three-step orbit transfer strategy to carry out the fuel consumption required by orbit transfer, and expressing the price as:

C _ij ＝Δv _hm +Δv _ym +Δv _tx

4. The method for online multi-to-multi-task planning on a distributed satellite according to claim 3, wherein the three-step orbit transfer strategy comprises three phases of huffman orbit transfer, out-of-plane orbit transfer and phase modulation maneuver, and the service spacecraft sequentially performs the huffman orbit transfer, out-of-plane orbit transfer and phase modulation maneuver, and the specific orbit transfer process comprises the following steps:

5. The method for online multi-to-multi-tasking planning on a distributed satellite according to claim 4, wherein updating the prices of the selected tasks relative to all the service spacecraft comprises the following specific procedures:

wherein i' noteqi ₀ ，

if there is no surplus second largest task j', then

/>

6. The method for online multi-to-multi-task planning on a distributed satellite according to claim 5, wherein the surplus score of the successful task is calculated by:

wherein ,

average value, n, representing surplus obtained by successful tasks in such a task allocation scheme _f Representing the number of failed tasks, p _t Representing an adjustment score for each failed task. />