CN109034670B - Satellite on-orbit activity planning method and system - Google Patents

Satellite on-orbit activity planning method and system Download PDF

Info

Publication number
CN109034670B
CN109034670B CN201811036402.1A CN201811036402A CN109034670B CN 109034670 B CN109034670 B CN 109034670B CN 201811036402 A CN201811036402 A CN 201811036402A CN 109034670 B CN109034670 B CN 109034670B
Authority
CN
China
Prior art keywords
activity
orbit
satellite
planning
activities
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201811036402.1A
Other languages
Chinese (zh)
Other versions
CN109034670A (en
Inventor
罗亚中
梁军
朱阅訸
牟帅
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
National University of Defense Technology
Original Assignee
National University of Defense Technology
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by National University of Defense Technology filed Critical National University of Defense Technology
Priority to CN201811036402.1A priority Critical patent/CN109034670B/en
Publication of CN109034670A publication Critical patent/CN109034670A/en
Application granted granted Critical
Publication of CN109034670B publication Critical patent/CN109034670B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
    • G06Q10/06315—Needs-based resource requirements planning or analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Economics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • General Physics & Mathematics (AREA)
  • Strategic Management (AREA)
  • Educational Administration (AREA)
  • Development Economics (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • Marketing (AREA)
  • General Business, Economics & Management (AREA)
  • Game Theory and Decision Science (AREA)
  • Operations Research (AREA)
  • Computer Hardware Design (AREA)
  • Evolutionary Computation (AREA)
  • Geometry (AREA)
  • General Engineering & Computer Science (AREA)
  • Supply And Distribution Of Alternating Current (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention discloses a method and a system for planning on-orbit activities of satellites, wherein the method comprises the following steps: 1) establishing a mathematical model considering practical engineering constraints, and mathematically describing the satellite in-orbit motion planning problem; 2) establishing a satellite on-orbit activity planning field knowledge model based on predicate logic and designing a group of priority heuristic rules by combining the field knowledge; 3) according to the guidance of the priority rule, a planning scheme meeting all constraint conditions is obtained through one-time independent discrete event simulation; 4) and randomly adjusting the weight of each influence factor in the priority rule, and iteratively searching a planning scheme with better goodness performance. The method can rapidly obtain a satisfactory planning scheme of the satellite in-orbit movement planning problem in a short time with very small calculation cost, gives consideration to high calculation efficiency and quality of planning solutions, and has the advantages of correct and reasonable solution method, rapid and effective calculation process, good applicability to actual engineering tasks and the like.

Description

Satellite on-orbit activity planning method and system
Technical Field
The invention relates to the technical field of satellite remote sensing, in particular to a priority rule-based fast heuristic method for solving the problem of satellite on-orbit activity planning.
Background
With the fact that geostationary orbit (hereinafter abbreviated as GEO) plays more and more important roles in the fields of communication, remote sensing, navigation and the like, it is of great significance to monitor GEO satellite beams by adopting low orbit (hereinafter abbreviated as LEO) satellite formation so as to grasp the use condition of frequency resources of the GEO satellite beams. In the GEO satellite beam monitoring task, successful execution of many operation instructions on the LEO satellite is required to ensure that the whole in-orbit task is successfully completed. Those basic operational instructions that can be directly executed on the LEO satellite are collectively referred to as satellite in-orbit activity. The process of arranging a plurality of in-orbit activities on the satellite on a time line to form a reasonable in-orbit activity execution scheme by taking a certain planning target as guidance is defined as the in-orbit activity planning of the satellite.
The satellite in-orbit activity planning can be regarded as a sub-problem in satellite in-orbit task planning and is a combined configuration operation research problem which is difficult to solve. Currently, there are precise algorithms and approximate algorithms for solving such problems. The precise algorithm has a branch-and-bound algorithm, which tries to find the optimal solution of the problem, but practice shows that the algorithms can only solve small-scale problems. To remedy this deficiency, some approximation algorithms have been proposed to solve the problem. The approximation algorithm is a modern evolution algorithm combined with certain improvement strategies, such as a differential evolution algorithm based on a constraint preprocessing strategy and an improved genetic algorithm based on a heuristic strategy, and the like, although the algorithms can obtain an approximately optimal solution of the problem, the calculation efficiency is low, the consumed calculation time is often intolerable, and the algorithms are not suitable for actual engineering requirements, so that new exploration is needed.
Disclosure of Invention
The invention particularly provides a satellite in-orbit activity planning method, which aims to overcome the defects of low calculation efficiency, long operation time and the like in the prior art.
In order to achieve the above object, the present invention provides a method for planning on-orbit activities of a satellite, comprising:
s1, establishing a real nonlinear programming model to describe the problem of satellite in-orbit activity programming;
s2, establishing a predicate logic-based satellite on-orbit activity planning field model, and establishing a priority heuristic rule according to the field model;
s3, obtaining a planning scheme meeting all constraint conditions in the real nonlinear programming model through one independent discrete event simulation according to the guidance of a priority heuristic rule;
and S4, randomly adjusting the weight of each influence factor in the prior heuristic rule, and iteratively searching a planning scheme with the goodness performance meeting the expected target.
In order to achieve the above object, the present invention further provides a satellite in-orbit activity planning system, which includes a processor and a memory, where the memory stores a satellite in-orbit activity planning program, and the processor executes the steps of the satellite in-orbit activity planning method when operating the satellite in-orbit activity planning program.
According to the method and the system for planning the in-orbit activities of the satellite, provided by the invention, a real nonlinear programming model is established according to the numerical constraints of in-orbit resources and the sequential logical constraints between the in-orbit activities, the problem of planning the in-orbit activities of the satellite is mathematically described, a domain model for planning the in-orbit activities of the satellite is established based on predicate logic, domain knowledge such as in-orbit activity expression and related constraint expression is accurately described, when the method and the system are oriented to actual engineering, the number and the types of in-orbit activities in the established domain model for planning the in-orbit activities of the satellite can be correspondingly adjusted according to actual tasks, and the method and the system have the advantage; designing a group of prior heuristic rules by combining with knowledge in the field of satellite in-orbit activity planning, and obtaining a planning scheme meeting all constraint conditions in a real nonlinear programming model through one-time independent discrete event simulation calculation under the guidance of the prior rules; on the basis, a strategy of randomly adjusting the weight coefficient is added to obtain a planning scheme with the goodness performance meeting the expected target; compared with the planning result of a modern evolutionary algorithm combined with a certain improvement strategy, the planning result of the method has higher quality of the planning solution and higher calculation speed, can quickly obtain a satisfactory planning scheme of the satellite in-orbit activity planning problem in a short time with very small calculation cost, has high calculation efficiency and quality of the planning solution, and has the advantages of correct and reasonable solution method, quick and effective calculation process, good applicability to actual engineering tasks and the like.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the structures shown in the drawings without creative efforts.
Fig. 1 is a schematic basic flowchart of a method for planning an in-orbit satellite activity according to an embodiment of the present invention;
fig. 2 is a flowchart illustrating a simulation of an independent discrete event in a method for planning an in-orbit satellite activity according to an embodiment of the present invention;
fig. 3 is a pseudo code of a method for planning an in-orbit satellite activity according to an embodiment of the present invention;
FIG. 4 is a graph comparing the calculated performance of two methods according to one embodiment of the present invention;
FIG. 5 is a graph comparing the calculated performance of the two methods according to the second preferred embodiment of the present invention;
FIG. 6 is a comparison graph of the calculated performance of the two methods in the third preferred embodiment of the present invention.
The implementation, functional features and advantages of the objects of the present invention will be further explained with reference to the accompanying drawings.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
It should be noted that all the directional indicators (such as up, down, left, right, front, and rear … …) in the embodiment of the present invention are only used to explain the relative position relationship between the components, the movement situation, etc. in a specific posture (as shown in the drawing), and if the specific posture is changed, the directional indicator is changed accordingly.
In addition, the descriptions related to "first", "second", etc. in the present invention are only for descriptive purposes and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. In the description of the present invention, "a plurality" means at least two, e.g., two, three, etc., unless specifically limited otherwise.
In the present invention, unless otherwise expressly stated or limited, the terms "connected," "secured," and the like are to be construed broadly, and for example, "secured" may be a fixed connection, a removable connection, or an integral part; the connection can be mechanical connection, electrical connection, physical connection or wireless communication connection; they may be directly connected or indirectly connected through intervening media, or they may be connected internally or in any other suitable relationship, unless expressly stated otherwise. The specific meanings of the above terms in the present invention can be understood by those skilled in the art according to specific situations.
In addition, the technical solutions in the embodiments of the present invention may be combined with each other, but it must be based on the realization of those skilled in the art, and when the technical solutions are contradictory or cannot be realized, such a combination of technical solutions should not be considered to exist, and is not within the protection scope of the present invention.
The invention provides a method and a system for planning on-orbit activities of a satellite.
The first embodiment is as follows:
as shown in fig. 1, an embodiment of the present invention provides a method for planning an in-orbit motion of a satellite, including the following steps:
s1, establishing a real nonlinear programming model according to the sequential logic constraint between the on-orbit resource value constraint and the on-orbit activity, and describing the satellite on-orbit activity programming problem;
in the planning problem of the on-orbit motion of the satellite, let M ═ A1,A2,…,Ak,…,AmDenotes a set of m on-track activities, BtRepresenting a set of activities performed in parallel on the satellite at time t, AlAny one activity sequence in the activity sequence set formed by the sequential logical relationship is represented. In order to minimize the total time span (Makespan) for all activities to complete execution, the start time t of each activity is chosenk(k ═ 1,2,. multidot.m) as setAnd (6) counting variables.
A real nonlinear programming model is established to describe the satellite in-orbit motion programming problem, and the mathematical expression of an objective function is as follows:
f＝min(Makespan)＝min[max(tk+Δtk)](k＝1,2,...,m) (1)
wherein, Δ tkIs an activity AkThe duration of execution.
Nonlinear programming means that the target function or constraint condition of the programming is not a linear function of the independent variable or is difficult to express by the linear function. Real nonlinear programming is nonlinear programming in the real domain.
The mathematical formulation of each constraint is as follows:
Figure BDA0001790976330000041
Figure BDA0001790976330000042
∑PSr≤Pavailable,r∈Bt(4)
wherein, PSrSet of activities B for parallel executiontMiddle activity ArConsumption of electrical power resources during execution, Σ PSrAs set BtSum of electric power resource consumptions of all activities in, PavailableThe amount of electrical power resources available on the satellite at the current time.
The constraint (2) indicates that the value of the start time of any activity in the activity set M cannot exceed the sum of the continuous execution times of all activities, so that the search space of the design variables (the start execution time of each activity) can be effectively reduced. Constraint (3) indicates that any two activities with sequential logical relationship constraints cannot overlap on the timeline, i.e., the latter activity must begin execution after its prior activity execution is complete. Constraint (4) indicates that at any one time the sum of the electric power resource consumptions of all activities performed in parallel cannot exceed the maximum available electric power resource quantity currently available on the satellite.
S2, establishing a predicate logic-based satellite on-orbit activity planning field model, and designing a set of priority heuristic rules according to the field model;
predicate logic is a form of symbolic logic system, belongs to a formal language system, is relatively close to natural language of human, is the earliest logic applied to artificial intelligence to express knowledge, and has been widely applied to knowledge engineering. The basic components of predicate logic are the predicate symbols, variable symbols, function symbols, and constant symbols, and are separated by circle brackets, square brackets, flower brackets, and commas to represent relationships between each other. For example, to represent "ROBOT (ROBOT) within Room number 1 (ROOM 1)", the predicate logic representation may be applied as:
INROOM(ROBOT,ROOM1)
in the formula, ROBOT and ROOM1 are constant symbols, and INROOM is a predicate symbol. Constant signs can be used to represent actual objects, or anything with a name; the variable symbol does not necessarily explicitly refer to which entity; the function notation represents a function within the domain, e.g. the function notation more may be used to represent a mapping between a person and his mother.
The domain model is the standardized definition and expression of main element concepts and interaction relations in the field of satellite in-orbit activity planning, and is the model basis for solving the satellite in-orbit activity planning problem.
Further, the S2 includes the following steps:
s201, establishing a predicate logic-based satellite in-orbit motion planning field model.
When modeling is carried out by using a knowledge representation method based on predicate logic, the satellite in-orbit activity planning field model can be represented by a quadruple.
OAPDKM＝＜Obj,Predicate,Function,Activity＞ (5)
Wherein Obj is a research object in the field of satellite in-orbit Activity planning, Predicate is a Predicate of each object in a satellite system and is used for describing a state attribute of the object, Function is various calculation functions used in the execution process of in-orbit Activity, and Activity is in-orbit Activity that each object can take, and the Predicate (namely, the state attribute) of the object is changed through the execution of in-orbit Activity.
The object Obj concept defines: in the knowledge model of the satellite in-orbit motion planning field, an object represents an element of interest in a satellite subsystem, and the element corresponds to the definition of each in-orbit motion of the satellite one by one, can be an actually existing device on the satellite, and can also be a virtual object. For the satellite on-orbit motion planning problem, the object types are defined as follows:
Obj＝{Id,Name,Subsystem} (6)
wherein, Id is the unique number of the object type, Name is the Name of the object type, and Subsystem is the satellite Subsystem to which the object type belongs.
Predicate concept definition: the state attribute of each object in the field of the satellite on-orbit activity planning is described by a predicate, and the value can be true or false. The formalization of the predicate is defined as follows:
Prediction＝{(P？O1-Obj1…？On-Objn)} (7)
where P is the predicate name, OiIs a predicate parameter, is an object instance, ObjiIs the object type to which the parameter belongs.
The Function concept defines: in the field of satellite in-orbit activity planning, functions refer to various calculation functions used, including activity duration function dur _ time, satellite system electric power resource value change function P _ value, other calculation functions other _ fun and the like. The formalization of the function is defined as follows:
Function＝{dur_time,P_value,other_fun} (8)
activity concept definition: the activity is the most core data structure in the knowledge model of the in-orbit activity planning field of the satellite, and refers to a basic operation instruction set which can be directly executed by each subsystem of the satellite. When the predicate logic is used for representing the activity, the activity is composed of a parameter list, duration, preconditions and execution effect, and formally defined as follows:
Activity＝{id,parameters(Oi-Obji),duration(dur_time),precondition(λ,Pc),effect(U,P_value)} (9)
wherein id is the activity number; parameters (O)i-Obji) Is a parameter list consisting of parameters of objects involved in the activity, (O)i-Obji) Is the ith object parameter in the list and the object type to which it belongs; duration (dur _ time) is a duration constraint for the activity; precondition (λ, P)c) Is a precondition for the execution of an activity that can only be activated if the fact of the system state satisfies this condition, λ is the set of excited states represented by predicates in the precondition, i.e. the state constraint, PcIs a resource constraint in the precondition; effect (U, P _ value) is the execution effect of the activity, and U refers to the system state of the predicate representation updated after the activity is executed.
In the precondition and the execution effect of the activity, the state of the predicate expression and the activity are constrained by a time relation. That is, the states in λ and U sets may be true at the start time (at start), end time (at end), or duration of the activity (over all). In the precondition of activity, the resource value constraint P is judged by adding predicate comparison operator (<, ≦ or ≧, ═ or ═ to)cWhether or not it is satisfied. In the execution effect of the activity, the numerical change P _ value of the system resource in the execution effect of the activity is described by adding predicate calculation formulas (+, -, /).
S202, a set of preferential heuristic rules is designed by combining with knowledge in the field of satellite on-orbit activity planning.
At a certain moment, all activities satisfying the state constraint λ are searched to form an alternative activity set (SetAlternative). Activity A in the alternative Activity setkPriority of Activity priority ofkIs defined as:
Figure BDA0001790976330000061
wherein the aggregate As is with AkThe activity sequence with the longest sum of duration in the activity sequences formed by the constraint (3) is satisfied;
Figure BDA0001790976330000062
is A in the aggregate AskThe sum of the durations of all subsequent activities of (a);
Figure BDA0001790976330000063
is A in the aggregate AskThe completion time of the prior activity of (a); PS (polystyrene) with high sensitivitykIs AkThe amount of electrical power resources occupied when executing; coefficient alpha1、α2And alpha3It is shown that the weights of the 4 product terms on the right side of the formula (10) are different in priority.
Priority rule 1: and performing descending order on the activities in the alternative activity set according to the priority, and preferentially arranging the activities with high priority on the time line.
Priority rule 2: when the arrangement is completed according to the priority rule 1 at a certain time, if the satellite system still has available electric power resources, the remaining activities in the alternative activity set may be arranged in ascending order according to the size of the consumption amount of the electric power resources. As long as the amount of electrical power resources remaining in the satellite system is still greater than the amount of electrical power resource consumption for the activities that are reordered further forward, the activities are arranged as far as possible in order to make full use of the electrical power resources P available at each moment in time from the satellite systemavailable。
Priority rule 3: the activities in the alternative activity set are all scheduled for execution as early as possible to achieve compactness of the arrangement of the activities on the timeline.
The searching movement operation of the activity is to search the corresponding activity according to three priority rules if the activity satisfies the electric power resource constraint P when executedcIt is arranged on the timeline of the current scheme, i.e.:
s(Plan)→SearchMove(Activity,Rule)→sN(PlanN) (11)
where s (plan) is the current scheme with some activities already deployed, SearchMove (Activity, Rule) is the search move operator, Rule is the above three priority rules, sN(PlanN) Is to search for the moving operatorActivity lays a new solution behind the original solution s (plan).
And S3, obtaining a planning scheme meeting all constraint conditions through one-time independent discrete event simulation according to the guidance of the priority rule in the S2.
As shown in FIG. 2, the detailed steps of a discrete event simulation independently include:
1) and (5) initially setting. Initializing the states represented by the system predicates, initializing the number of available electric power resources P on the satelliteavailableAll modeled on-track activity information is entered to form the set of unexecuted activities setuntork.
2) And (5) predicate logic reasoning. At the simulation time t, if the excitation state λ represented by the predicate in the precondition of an activity in the set setuntwork is consistent with the current state represented by the predicate in the satellite system (that is, the values of the two predicates are both true or false), the activity is excited and selected into the alternative activity set SetAlternative.
3) And searching for a moving operation. Under the guidance of the three priority rules, the search movement operator defined by the formula (11) arranges the activities in the set SetAlternative on a time line and occupies a corresponding amount of electric power resources, and the starting execution time of the search movement operator is the simulation time t. These activities are moved into the executing activity set SetWorking and deleted from the sets setUnwork and setAlternativative.
4) And updating system predicates. And if the completion time of the activity in the set SetWorking is the simulation time t, releasing the electric power resource occupied by the activity during the execution, and updating the state represented by the system predicate corresponding to the activity. These activities are moved into the set of completed activities SetWorked and deleted from the set SetWorking.
5) Simulating propulsion and calculating step length. At the simulation time t, if the state represented by the system predicate is updated, which indicates that there is an activity that completes execution and releases the corresponding electric power resource in the set SetWorking, the activity may be rearranged at this moment without advancing the simulation time (i.e. the simulation step size SimStep ═ 0). If the system predicates are not updated, and the activities which are executed at the moment are fullElectrical power resources, then the campaign cannot be re-arranged at this point, requiring the simulation time to be advanced to the earliest completion time of the campaign being executed in the set SetWorking (i.e., simstart { (t) } min { (t)c+Δtc)|c∈SetWorking}-t)。
6) And (6) outputting a planning scheme. And 5) continuously advancing simulation according to the step 5) until all activities in the SetUnwork are arranged completely, and outputting a planning scheme meeting all constraint conditions.
Thus, an independent discrete event simulation is completed.
And S4, randomly adjusting the weight of each influence factor in the formula (10), and iteratively searching a planning scheme with better goodness performance.
In order to obtain a planning scheme with better goodness performance, an iterative optimization process is added on the basis of one discrete event simulation, as shown in fig. 3. In the iterative optimization process, the sum of the durations of the subsequent activities in the formula (10) is adjusted in a random manner
Figure BDA0001790976330000081
Completion time of prior activity
Figure BDA0001790976330000082
Duration of the activity itself Δ tkAmount of electric power resources consumed by the activity PSkThe weighting of the four key influencing factors in priority rule 1, i.e. for the coefficient α1、α2And alpha3A random assignment strategy is adopted. The detailed steps of iterative optimization include:
1) and (5) initially setting. And setting the maximum iteration times of the overall optimization and the local optimization. The initial set minimum completion time span MinMakespan is the sum of the durations of all activities to be planned.
2) And (4) overall optimization. As long as the maximum number of iterations of the global optimization has not been reached, the coefficient α is then aligned by equation (12)1、α2And alpha3In the interval [0,1]Internally and randomly assigning values, and obtaining a coefficient alpha at the moment through one discrete event simulation1、α2And alpha3Correspond to each otherAnd a completion time span Makespan for the plan. If the Makespan at the moment is smaller than Minmakespan, assigning the Makespan at the moment to Minmakespan, and recording the planning scheme and the coefficient alpha at the moment1、α2And alpha3The value of (c).
{α1,α2,α3}～U(0,1) (12)
3) Local optimization. Obtaining the coefficient alpha after the integral optimization1、α2And alpha3On the basis of the value of (3), as long as the maximum iteration number of local optimization is not reached, random numbers are increased or decreased within a small range of 0-0.1, which is similar to the overall optimization method in the step 2). After the local optimization is completed, a minimum Makespan planning scheme and a coefficient alpha are obtained1、α2And alpha3The value of (c).
And finally, completely finishing the rapid solving process of the satellite in-orbit activity planning problem by using a heuristic method based on a priority rule.
Preferred embodiment 1
As shown in fig. 1, the heuristic method for rapidly planning the in-orbit activity of a satellite in this embodiment includes:
s00, inputting parameters, and inputting scene parameters of the LEO satellite in-orbit activity planning in the following GEO satellite beam monitoring tasks:
table 1 lists the relevant attributes of 32 in-orbit activities in the seven major subsystems of the LEO satellite, such as the duration of the activity Δ t, the amount of consumed electric power resources PS while the activity is performed, the amount of released electric power resources PE while the activity is completed, and its a priori activity number idPr. And assuming that the maximum power consumption of the LEO satellite system for providing the electric power resource is 8.5W.
S01, establishing a mathematical model considering practical engineering constraints, and mathematically describing the satellite in-orbit activity planning problem;
the on-orbit motion planning problem is described by the mathematical model established by the above equations (1) to (4).
S02, establishing a predicate logic-based satellite on-orbit activity planning field model and designing a set of priority heuristic rules by combining field knowledge;
and (5) substituting the 32 in-orbit activities in the S00 into the above equations (5) to (9) to obtain a satellite in-orbit activity planning field model.
S03, obtaining a planning scheme meeting all constraint conditions through one-time independent discrete event simulation according to the guidance of priority rules;
and substituting the 32 modeled on-orbit activities in the S02 into a discrete event simulation process to obtain a planning scheme.
S04, randomly adjusting the weight of each influence factor in the priority rule, and iteratively searching a planning scheme with better goodness performance;
the coefficient α is randomly changed by the above equation (12)1、α2And alpha3And substituting the one-time discrete event simulation in the step S3 into the iterative optimization process until a planning scheme with better goodness performance is obtained.
Table 1 example one 32 in-orbit active sets
Figure BDA0001790976330000091
In this embodiment, the calculation result is as follows:
table 2 values of the objective function in example one
Figure BDA0001790976330000092
In order to verify the goodness performance of the calculation results shown in table 2, the same solution was performed on this embodiment by using a differential evolution algorithm combined with a constrained preprocessing strategy on the same computer. Fig. 4 shows the evolution curve of the planning objective function value Makespan with the method operation time in the method of the present invention and the global optimization method. Fig. 4(a) shows that the method of the present invention has Makespan 972s planning scheme when it runs to 3 seconds, and fig. 4(b) shows that the global optimization method has Makespan 972s planning scheme when it runs to 221.3 seconds. And the global optimization method starts to obtain a feasible solution meeting all the constraint conditions at the 137.4 th second, and starts to converge to the optimal solution of Makespan 928.041681s at the 823.3 th second. This shows that on the premise of obtaining a planning scheme with equal goodness performance, the operation time of the heuristic method is about 1/73.77 of that of the global optimization method, and the calculation efficiency is far higher than that of the global optimization method. Meanwhile, the method can give consideration to high calculation efficiency and optimal performance of planning solution, and is suitable for actual engineering tasks.
The second preferred embodiment:
the present embodiment is basically the same as the first embodiment in terms of the steps, and the main difference is that the initial parameters input in step S00 are different, and other steps are the same as the first embodiment.
And S00, inputting parameters, and inputting the following extended scene parameters of the on-track activities:
table 3 extends again 29 new in-orbit activities compared to table 1, for a total of 61 in seven major subsystems of the satellite. At this time, the satellite system may provide a maximum usage power consumption of 12W of the electric power resource.
TABLE 3 second embodiment 61 in-orbit active sets
Figure BDA0001790976330000101
The steps of S01-S04 are exactly the same as those of the first embodiment.
In this embodiment, the calculation result is as follows:
TABLE 4 values of the objective function in example two
Figure BDA0001790976330000111
Similarly, the same solution is performed on the embodiment by using a differential evolution algorithm combined with a constraint preprocessing strategy on the same computer. Fig. 5 shows the evolution curves of the planning objective function values Makespan of the two methods in the embodiment along with the operation time of the method. Fig. 5(a) shows that the method of the present invention has obtained Makespan 1513s planning scenario when it runs to the 4 th second, while fig. 5(b) shows that the global optimization method has obtained Makespan 1513s planning scenario when it runs to the 2873.6 th second, and starts to converge to the optimal solution of Makespan 1468.08370s at the 27736.6 th second. This shows that on the premise of obtaining a planning scheme with equal goodness performance, the operation time of the heuristic method is about 1/718.40 of that of the global optimization method, and the calculation efficiency is far higher than that of the global optimization method.
The third preferred embodiment:
the present embodiment is basically the same as the first embodiment in terms of the steps, and the main difference is that the initial parameters input in step S00 are different, and other steps are the same as the first embodiment.
And S00, inputting parameters, and inputting scene parameters of the on-orbit activities of the following expansion subsystems:
compared with table 3, table 5 adds a new subsystem, and continues to extend 19 in-orbit activities, for a total of 80 in-orbit activities out of eight major subsystems of the satellite. The maximum power consumption of the satellite system for providing the electric power resource is still 12W.
TABLE 5 example three 80 in-orbit active sets
Figure BDA0001790976330000112
Figure BDA0001790976330000121
The steps of S01-S04 are exactly the same as those of the second embodiment.
In this embodiment, the calculation result is as follows:
TABLE 6 values of the objective function in EXAMPLE III
Figure BDA0001790976330000122
Similarly, the same solution is performed on the embodiment by using a differential evolution algorithm combined with a constraint preprocessing strategy on the same computer. Fig. 6 shows the evolution curves of the planning objective function values Makespan of the two methods in the embodiment along with the operation time of the method. Fig. 6(a) shows that the method of the present invention has obtained Makespan 2156s of planning solution when it runs to 6 th second, and fig. 6(b) shows that the global optimization method has obtained Makespan 2156s of planning solution when it runs to 7292.3 th second, and starts to converge to Makespan 2137.89347s of optimal solution at 46746.8 th second. This shows that on the premise of obtaining a planning scheme with equal goodness performance, the operation time of the heuristic method is about 1/1215.38 of that of the global optimization method, and the calculation efficiency is far higher than that of the global optimization method.
Example two
The embodiment of the invention also provides a satellite in-orbit activity planning system which comprises a processor and a memory, wherein the memory stores a satellite in-orbit activity planning program, and the processor executes the steps of the satellite in-orbit activity planning method when operating the satellite in-orbit activity planning program. The implementation of the system refers to the implementation of the embodiment of the satellite in-orbit activity planning method.
The above description is only a preferred embodiment of the present invention, and the protection scope of the present invention is not limited to the above embodiments, and all technical solutions belonging to the idea of the present invention belong to the protection scope of the present invention. It should be noted that various modifications and adaptations to those skilled in the art may be made without departing from the principles of the present invention.

Claims (8)

1. A method for planning the in-orbit motion of a satellite is characterized by comprising the following steps:
s1, establishing a real nonlinear mathematical model according to the sequential logic constraint between the on-orbit resource value constraint and the on-orbit activity, and describing the planning problem of the on-orbit activity of the satellite;
s2, establishing a predicate logic-based satellite on-orbit activity planning field model, and designing a set of priority heuristic rules according to the field model;
s3, obtaining a planning scheme meeting all constraint conditions in the real nonlinear mathematical model through one independent discrete event simulation according to the guidance of a priority heuristic rule;
and S4, randomly adjusting the weight of each influence factor in the prior heuristic rule, and iteratively searching a planning scheme with the goodness performance meeting the expected target.
2. The method for planning the in-orbit satellite activity according to claim 1, wherein the step S1 comprises the steps of:
s101, the mathematical expression of the objective function in the established real nonlinear programming model is as follows:
Figure FDA0002611125050000012
where Makespan represents a set of M on-track activities (M ═ { a ═ a)1,A2,…,Ak,…,AmTime span when all on-track activities are completed; Δ tkFor any one on-orbit activity A in the set MkThe duration of (d); t is tk(k 1, 2.. times.m) is a planning variable representing any one of the on-orbit activities a in the set MkThe start execution time of (c);
s102, in the real nonlinear programming model, a first constraint condition is that the value of the starting execution time of any activity in the set M cannot exceed the sum of the duration times of all on-orbit activities in the set M:
Figure FDA0002611125050000011
s103, in the real nonlinear programming model established, the second constraint condition is that any two on-orbit activities constrained by a sequential logical relationship cannot be overlapped on a time line, and the latter on-orbit activity can be executed only after the prior activity is completed:
tj-ti≥Δti,(i,j)∈Al
k≥j＞i (3)
wherein A islRepresenting any activity sequence in an activity sequence set formed by sequential logical relations;
s104, in the real nonlinear programming model established, the third constraint condition is that at any time t, the sum of the electric power resource consumptions of all in-orbit activities executed in parallel does not exceed the current available electric power resource value on the satellite:
∑PSr≤Pavailable,r∈Bt(4)
wherein, BtRepresenting a set of all in-orbit activities, PS, performed in parallel on the satellite at time trAs set BtAny one of the on-orbit activities ArConsumption of electrical power resources during execution, Σ PSrAs set BtSum of electric power resource consumptions, P, of all on-orbit activities performed in parallelavailableThe amount of electrical power resources available on the satellite for time t.
3. The satellite in-orbit activity planning method according to claim 1 or 2, wherein the step S2 comprises the steps of:
s201, the predicate logic-based satellite in-orbit activity planning domain model can be represented by the following four-tuple:
OAPDKM＝＜Obj,Predicate,Function,Activity＞ (5)
wherein Obj represents a research object in the field of satellite in-orbit Activity planning, Predicate represents a Predicate of each research object in a satellite system and is used for describing a state attribute of the object, Function represents various calculation functions used in the execution process of in-orbit Activity, Activity represents in-orbit Activity that each research object can adopt, and the state attribute of the object is changed through the execution of the in-orbit Activity;
the object type is defined as follows:
Obj＝{Id,Name,Subsystem} (6)
the method comprises the following steps that Id is the unique number of an object type, Name is the Name of the object type, and Subsystem is the satellite Subsystem to which the object type belongs;
the formalization of the predicate is defined as follows:
Prediction＝{(P？O1-Obj1…？On-Objn)} (7)
where P is the predicate name, OiIs a predicate parameter, is an instance of an object type, ObjiThe type of the object to which the parameter belongs;
the formalization of the function is defined as follows:
Function＝{dur_time,P_value,other_fun} (8)
wherein dur _ time is a duration function of on-orbit activity, P _ value is a numerical value change function of electric power resources of the satellite system, and other _ fun is other calculation functions;
the formalization of the on-track activity is defined as follows:
Activity＝{id,parameters(Oi-Obji),duration(dur_time),precondition(λ,Pc),effect(U,P_value)} (9)
wherein id is an on-orbit activity number; parameters (O)i-Obji) A parameter list formed by object parameters related to the on-orbit activity; (O)i-Obji) The parameter of the ith object in the list and the object type of the ith object are obtained; duration (dur _ time) is a duration constraint for on-track activity; precondition (λ, P)c) A precondition for execution of an on-track activity; lambda is a set of excitation states represented by predicates in the precondition; pcIs a resource constraint in the precondition; effect (U, P _ value) is the execution effect of the on-track activity; u is the updated system state represented by the predicate after the on-orbit activity is executed;
s202, searching all activities meeting the state constraint lambda at the moment t to form an alternative activity set (SetAlternative), wherein the on-orbit activity A in the alternative activity setkPriority of Activity priority ofkIs defined as:
Figure FDA0002611125050000031
wherein the aggregate As represents and AkThe activity sequence with the longest sum of duration time in the activity sequences formed by satisfying the constraint of S103;
Figure FDA0002611125050000032
represents A in the set AskThe sum of the durations of all subsequent activities of (a);
Figure FDA0002611125050000033
is A in the aggregate AskThe completion time of the prior activity of (a); PS (polystyrene) with high sensitivitykIs AkThe amount of electrical power resources occupied when executing; alpha is alpha1、α2And alpha3Is a weight coefficient; (ii) a
S203, performing descending order arrangement on the on-orbit activities in the alternative activity set SetAlternative according to the priority, and preferentially arranging the activities with high priority on the time line;
s204, when the arrangement is finished according to the priority rule of S203 at a certain moment and the satellite system still has available electric power resources, arranging the rest on-orbit activities in the alternative activity set SetAlternative in an ascending order according to the consumption of the electric power resources;
s205, arranging all on-orbit activities in the alternative activity set SetAlternative to be executed at the earliest possible time;
s206, searching corresponding activities according to the priority rules of S203-S205, and satisfying the electric power resource constraint P when the activities are executedcWhen the activity is scheduled on the timeline of the current scenario, i.e.:
s(Plan)→SearchMove(Activity,Rule)→sN(PlanN) (11)
wherein S (plan) represents the current scheme with activities already arranged, SearchMove (Activity, Rule) represents search mobile operator, Rule represents priority rules of S203-S205, SN(PlanN) Represents a new scheme obtained by arranging the new Activity obtained by searching the mobile operator after the original scheme s (plan).
4. The method for planning an in-orbit satellite activity according to claim 3, wherein in the step S201, the state of the predicate representation and the time relation constraint between the activity are in the precondition and the execution effect of the activity; in the precondition of activity, the resource numerical value constraint P is judged by adding predicate comparison operatorcWhether satisfied, in the execution effect of the activity, by adding predicatesThe formula describes the value change P _ value of the system resource in the activity execution effect.
5. The method for planning the in-orbit satellite activity according to claim 3, wherein the step S3 comprises the steps of:
s301, initializing the state represented by system predicates and the available electric power resource quantity P on the satelliteavailableInputting all modeled on-track activity information to form an unexecuted activity set SetUnwork;
s302, at simulation time t, when an excitation state lambda represented by a predicate in a precondition of an activity in a set SetUnwork is consistent with a current state represented by the predicate in a satellite system, the activity is selected into an alternative activity set SetAlternative;
s303, under the guidance of the priority rules of S203-S205, arranging the activities in the set SetAlternativative on a time line by using a search mobile operator defined by the formula (11) and occupying electric power resources with corresponding values, wherein the starting execution time is a simulation time t, the starting execution time is shifted into the executing activity set SetWorking, and the starting execution time is deleted from the sets SetUnwork and SetAlternativative;
s304, when the completion time of the activity in the set SetWorking is the simulation time t, releasing the electric power resource occupied by the activity, updating the state represented by the system predicate corresponding to the activity, moving the activities into the completed activity set SetWorked, and deleting the activities from the set SetWorking;
s305, at a simulation time t, when the state represented by the system predicate is updated, the simulation time does not need to be advanced, namely the simulation step length SimStep is 0; when the system predicate is not updated, the simulation time is pushed to the earliest completion time of the activity being executed in the set SetWorking, i.e. SimStep is min { (t)c+Δtc)|c∈SetWorking}-t；
And S306, continuously advancing the simulation according to the S305 until all the activities in the SetUnwork are arranged completely, and outputting a planning scheme meeting all the constraint conditions.
6. The method for planning the in-orbit satellite activity according to claim 5, wherein the step S4 includes:
randomly adjusting the sum of the durations of the subsequent activities in equation (10)
Figure FDA0002611125050000041
Completion time of prior activity
Figure FDA0002611125050000042
Duration of the activity itself Δ tkAmount of electric power resources consumed by the activity PSkThe weights of the four key influencing factors in the precedence rule of step S203, i.e. the pair coefficient alpha1、α2And alpha3A random assignment mode is adopted, and the iterative optimization searching step comprises the following steps:
s401, setting the maximum iteration times of integral optimization and local optimization, and initially setting the minimum completion time span MinMakespan as the sum of the duration time of all on-track activities in the set SetUnwork;
s402, when the maximum iteration number of the whole optimization is not reached, the coefficient alpha is subjected to the equation (12)1、α2And alpha3In the interval [0,1]Internally and randomly assigning values, and obtaining a coefficient alpha at the moment through one discrete event simulation1、α2And alpha3A corresponding planning scheme and the completion time span Makespan of the scheme, when the Makespan at the moment is less than the MinMakespan, assigning the Makespan at the moment to the MinMakespan, and recording the planning scheme and the coefficient alpha at the moment1、α2And alpha3The value of (c):
{α1,α2,α3}～U(0,1) (12)
s403, obtaining the coefficient alpha after the integral optimization is completed1、α2And alpha3On the basis of the value, when the maximum iteration number of local optimization is not reached, the alpha is aligned within a small range of 0-0.11、α2And alpha3Increase or decrease random number, and obtain the coefficient alpha by one discrete event simulation1、α2And alpha3A corresponding planning scheme and a completion time span Makespan of the scheme, when the Makespan at the moment is smaller than the MinMakespan, assigning the Makespan at the moment to the MinMakespan to obtain a minimum planning scheme and a minimum coefficient alpha of the Makespan1、α2And alpha3The value of (c).
7. The method for planning the in-orbit satellite activity according to claim 6, wherein in step S4, the coefficient α in equation (12)1、α2And alpha3And obeying uniform distribution between 0 and 1.
8. A satellite in-orbit motion planning system, which comprises a processor and a memory, wherein the memory stores a satellite in-orbit motion fast planning program, and the processor executes the steps of the method according to any one of claims 1 to 7 when running the satellite in-orbit motion fast planning program.
CN201811036402.1A 2018-09-06 2018-09-06 Satellite on-orbit activity planning method and system Active CN109034670B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201811036402.1A CN109034670B (en) 2018-09-06 2018-09-06 Satellite on-orbit activity planning method and system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201811036402.1A CN109034670B (en) 2018-09-06 2018-09-06 Satellite on-orbit activity planning method and system

Publications (2)

Publication Number Publication Date
CN109034670A CN109034670A (en) 2018-12-18
CN109034670B true CN109034670B (en) 2020-10-27

Family

ID=64624330

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201811036402.1A Active CN109034670B (en) 2018-09-06 2018-09-06 Satellite on-orbit activity planning method and system

Country Status (1)

Country Link
CN (1) CN109034670B (en)

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112529241B (en) * 2020-09-18 2024-06-14 北京空间飞行器总体设计部 Remote sensing satellite cost-effectiveness balance optimization method
CN112883498B (en) * 2021-03-25 2022-04-19 中国人民解放军国防科技大学 On-orbit demand traction space station logistics supply planning method
CN115392813B (en) * 2022-10-31 2023-01-24 中国人民解放军国防科技大学 On-orbit event arrangement and resource allocation planning method based on hierarchical optimization
CN115828646B (en) * 2023-02-21 2023-04-25 中国人民解放军国防科技大学 Space station operation short-term scheme adjustment strategy simulation method and device under emergency condition
CN116011792B (en) * 2023-02-21 2023-06-27 中国人民解放军国防科技大学 Task time logic constraint reasoning method and device based on constraint hierarchical network
CN117312722B (en) * 2023-09-21 2024-03-19 华东师范大学 Multi-system task planning method based on SMT solver

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104361234A (en) * 2014-11-15 2015-02-18 北京理工大学 Method for optimizing multi-star multitask observation dispatching under complicated constraint condition
CN106384169A (en) * 2016-09-22 2017-02-08 合肥工业大学 Hyper-heuristic algorithm-based satellite task planning method

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104361234A (en) * 2014-11-15 2015-02-18 北京理工大学 Method for optimizing multi-star multitask observation dispatching under complicated constraint condition
CN106384169A (en) * 2016-09-22 2017-02-08 合肥工业大学 Hyper-heuristic algorithm-based satellite task planning method

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
空间目标多星多角度实时近距离三维详查研究;许丹丹等;《上海航天》;20180825;第35卷(第4期);第70-77页 *

Also Published As

Publication number Publication date
CN109034670A (en) 2018-12-18

Similar Documents

Publication Publication Date Title
CN109034670B (en) Satellite on-orbit activity planning method and system
Drake et al. Recent advances in selection hyper-heuristics
Srivatsava et al. Optimal test sequence generation using firefly algorithm
Jarboui et al. A combinatorial particle swarm optimization for solving multi-mode resource-constrained project scheduling problems
Lova et al. An efficient hybrid genetic algorithm for scheduling projects with resource constraints and multiple execution modes
Wu et al. Ensemble of metaheuristic and exact algorithm based on the divide-and-conquer framework for multisatellite observation scheduling
Sinha et al. Qubit routing using graph neural network aided Monte Carlo tree search
Basset et al. Compositional strategy synthesis for stochastic games with multiple objectives
Yu et al. An extended flexible job shop scheduling model for flight deck scheduling with priority, parallel operations, and sequence flexibility
Tarkesh et al. Facility layout design using virtual multi-agent system
Han et al. Multiple GEO satellites on-orbit repairing mission planning using large neighborhood search-adaptive genetic algorithm
Chen et al. Deep reinforcement learning for agile satellite scheduling problem
Yu et al. Combined Petri net modelling and AI-based heuristic hybrid search for flexible manufacturing systems—Part II. Heuristic hybrid search
Lin et al. When architecture meets AI: A deep reinforcement learning approach for system of systems design
US20230274216A1 (en) Delivery plan generation apparatus, delivery plan generation method, and program
US7139741B1 (en) Multi-objective optimization method
CN111190711B (en) BDD combined heuristic A search multi-robot task allocation method
Galván-López et al. Heuristic-based multi-agent monte carlo tree search
Undurti et al. A decentralized approach to multi-agent planning in the presence of constraints and uncertainty
Bettemir Optimization of time-cost-resource trade-off problems in project scheduling using meta-heuristic algorithms
Clarkson et al. Signposting for Design Process Improvement: A dynamic approach to design process planning
Karaman et al. Effective task assignment for complex uav operations using genetic algorithms
Theile et al. Learning to recharge: UAV coverage path planning through deep reinforcement learning
Zhang et al. Himacmic: Hierarchical multi-agent deep reinforcement learning with dynamic asynchronous macro strategy
CN115686031A (en) AUV (autonomous Underwater vehicle) informational path planning method based on learning and sampling

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant