CN108335012A - A kind of intelligence remote sensing satellite stratification distributed freedom cotasking planning system - Google Patents
A kind of intelligence remote sensing satellite stratification distributed freedom cotasking planning system Download PDFInfo
- Publication number
- CN108335012A CN108335012A CN201711433054.7A CN201711433054A CN108335012A CN 108335012 A CN108335012 A CN 108335012A CN 201711433054 A CN201711433054 A CN 201711433054A CN 108335012 A CN108335012 A CN 108335012A
- Authority
- CN
- China
- Prior art keywords
- task
- star
- satellite
- scheduling
- intelligent
- 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.)
- Pending
Links
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
Landscapes
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Economics (AREA)
- Operations Research (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Marketing (AREA)
- Educational Administration (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- Physics & Mathematics (AREA)
- General Business, Economics & Management (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Radio Relay Systems (AREA)
Abstract
The invention discloses a kind of intelligent remote sensing satellite stratification distributed freedom cotasking planning systems, it includes scheduler on more star task dispatchers and star, task in set of tasks to be allocated is distributed to more intelligent satellites having under its command and the Meta task that task is processed into scheduler Direct Recognition on star by more star task dispatchers, each intelligent satellite carries out United Dispatching using scheduler on its star to assigned new task and existing task, wherein, more star task dispatchers are before carrying out task distribution, pre-estimate the scheduling result of scheduler on related star, and in this, as the foundation of task distribution.The present invention successfully tackled key problems the prediction of imaging task schedulability, the more resource dynamic dispatchings of multitask under higher dimensional space, fining planning and scheduling algorithm design key scientific problems and technical barrier, so that the autonomous cotasking planning technology of more stars is preferably applied for national defense construction field, and promotes the rationalization of allocation plan and high efficiency that resource uses.
Description
Technical field
The present invention relates to remote sensing satellite technical field more particularly to a kind of intelligent remote sensing satellite stratification distributed freedom associations
Same task grouping.
Background technology
Remote sensing satellite is used as the artificial satellite of outer space remote sensing platform.In general, remote sensing satellite can orbit
Several years.Satellite orbit can determine as needed.Remote sensing satellite can cover the entire earth or specified times before the deadline
What region, when being run along geostationary orbit, it can continuously ground-to-ground ball surface specifies region to carry out remote sensing.All is distant
Sense satellite has been required for Remote Sensing Ground Station, and the satellite data that platform obtains from remote sensing fairground can monitor agricultural, forestry, sea
Situations such as ocean, territory, environmental protection, meteorology, remote sensing satellite mainly have meteorological satellite, Landsat and seasat three types.
Remote sensing satellite technology achieves larger progress at present, and the operating mode of different remote sensing satellites and use constraint are very
Generally there is complexity relatively independent task grouping, current remote sensing satellite technology to still have problems with:
(1) mission planning the complex nature of the problem.Intelligent satellite mission planning is in task, resource, constraint and optimization aim etc.
It is difficult to solve with optimization method that four aspects, which have certain particularity, common resource dispatching model,.
(2) complexity of dispatching algorithm and uncertainty.The randomness of dispatching algorithm, which makes scheduling result also, to be had not really
It is qualitative, while also increasing the difficulty of schedulability prediction.
(3) complexity of task samples selection.Different satellites can accumulate a large amount of historic task number during in orbit
According to how selecting typical representative sample to have certain difficulty to improve the execution efficiency of prediction algorithm.
(4) complexity of sample characteristics extraction.Imaging task typically is provided with attributive character of both static and dynamic:It is quiet
State attribute is mainly the association attributes for not changing with place set of tasks and changing that task independently has, such as the number of imaging task
According to type, resolution ratio, priority, demand observation duration, meteorological condition and imaging pattern etc.;Dynamic attribute collects with where task
The variation of conjunction and change, such as resource contention situation between description task, observation airplane can conflict situations.How in each generic attribute
It is equally sufficiently complex that select has the feature of decisive influence for prediction process.
(5) the imaging observation demand that the numerous users of different industries submit daily will break through tens thousand of, how carry multi-user
The a large amount of imaging tasks handed over are efficiently assigned to different remote sensing satellites and still lack effective theory and technology support.
(6) remote sensing satellite mission planning have complicated constraint, it is difficult to predict status information and complicated demand kind
Class so that satellite task planning problem is always the difficult point in system engineering field.
(7) sharply increasing with remote sensing satellite and imaging task number, how the high-resolution for obtaining New Satellite
Remote sensing image is quickly transferred to terrestrial user (i.e. the more earth station's scheduling of multitask) in time, still lacks effective technical support.
Invention content
In order to solve the problems in the prior art, the object of the present invention is to provide a kind of intelligent remote sensing satellite stratification distributions
The autonomous cotasking planning system of formula, imaging task schedulability of successfully having tackled key problems prediction, the multitask under higher dimensional space
Key scientific problems and the corresponding technical barriers such as more resource dynamic dispatchings, fining planning and scheduling algorithm design, make more stars
Autonomous cotasking planning technology can be preferably applied for actual national defense construction field, promote the reasonable of allocation plan
Change, it is more efficient to promote resource use.
To achieve the goals above, the technical solution adopted by the present invention is:
A kind of intelligence remote sensing satellite stratification distributed freedom cotasking planning system comprising more star task dispatchers
With scheduler on star, the task in set of tasks to be allocated is distributed to more intelligence having under its command and defended by more star task dispatchers
Star and the Meta task that task is processed into scheduler Direct Recognition on star, each intelligent satellite is using scheduler on its star to being assigned
New task and existing task carry out United Dispatching, wherein more star task dispatchers are before carrying out task distribution, in advance
The scheduling result of scheduler on the related star of estimation, and in this, as the foundation of task distribution, later stage scheduling result is avoided to feed back
Hysteresis quality.
Preferably, the set of tasks in rolling window is distributed to more intelligence having under its command and defended by more star task dispatchers
Star, each intelligent satellite is scheduled assigned new task and existing task using scheduler on its star, in current scrolling window
The initial time of mouth, more star task dispatchers are updated mission bit stream, delete times completed in a upper rolling window
Business and in the task that the initial time is carrying out, and by unappropriated task in a upper rolling window and upper one
The new task reached in rolling window is combined into the set of tasks in current scrolling window, and more star task dispatchers should
Set of tasks is allocated to the more intelligent satellites, wherein determines the starting of rolling window based on mixing triggering pattern
Moment, on the one hand, roll distribution every a period triggering, which is constant or moves according to preset rules
State changes;On the other hand, there is the event for making system mode change or triggering rolling distribution when by manual intervention.
Preferably, on the one hand, the period is arranged according to the observing and controlling period;On the other hand, described to make system mode
The event of change includes:Emergent observation mission is received, and the unallocated emergent observation mission accumulated reaches five either institutes
State the 5% of the intelligent satellite number that more star task dispatchers have under its command.
Preferably, more star task dispatchers include earth station and geostationary orbit telecommunication satellite, in the observing and controlling period
Within, the earth station carries out task distribution;Except the observing and controlling period, the geostationary orbit telecommunication satellite carries out task point
Match, and the emergent observation mission is generated by the intelligent satellite, wherein the geostationary orbit telecommunication satellite only into
The intelligent satellite with communications loop carries out task distribution therewith when row distribution, at this point, what more star task dispatchers had under its command
Intelligent satellite number refers to the quantity for the intelligent satellite for having communications loop with the geostationary orbit telecommunication satellite.
In the present invention, task distribution does not have to be with practical transmission executes in real time.That is, more stars are appointed
Being engaged in coordinator can at some time point, to all satellites progress task distribution having under its command, but only in real time by correlation times
Business is sent to the current intelligent satellite with communications loop, as the intelligent satellite for not having communications loop temporarily, then next
The task of distribution is sent on star scheduler in a call duration time window.Similarly, scheduling knot of the distributed task on star
Fruit nor is it necessary that feeds back to more star coordinators in real time.Even, scheduler is not after receiving newly assigned task on star
It is scheduled.This is conducive to arrange resource and work plan on star on star.Improve the predictability of corporate plan.
Preferably, the task scheduling strategy of scheduler is as follows on the star of each intelligent satellite:
(1) next week is generated using the complete rescheduling strategy in gradual method in the scheduling instance point of T- drivings
New task plan in phase time interval, the scheduling instance point of T- drivings are determined specifically according to given time interval T
Scheduling time point lT, 0≤l≤L, LT≤H < (L+1) T often reach a scheduling time point lT, then calculate and generate latter scheduling
The task scheduling in section [lT, (l+1) T], wherein l are positive integer, and T is given time interval, and L is maximum T- driving scheduling time
Number, H are total activation section,
(2) in C*The readjustment degree moment point of driving works as satellite using the operation plan correcting strategy in revision formula method
When operating in given scheduling interval, if at a time t (0 < t < H), the emergent observation mission cumulant C on startIt is super
Cross given threshold value C*When, then execute the calculating of readjustment degree, wherein threshold value C*For meet an urgent need observation mission critical integral number,
In addition to above two scheduling instance point, it is not scheduled in any other moment point,
It is as follows in the dispatching algorithm of the scheduling instance point of T- drivings:
Input:
The emergent observation mission set that arrived and be not scheduled before T- drives scheduling instance point;
The routine observation set of tasks for having received and not being scheduled before T- drives scheduling instance point;
Output:
-- the operation plan in subsequent time period T;
It is as follows:
Step 11 respectively fromWithWhether middle access time window falls into the routine observation in next period of time T
Task and emergent observation mission generate the routine observation set of tasks for waiting for that scheduling solvesWith emergent observation mission set
Step 12 willWithIt is integrated into an observation mission set;
Step 13 is ranked up the task in the observation mission set after integration according to the heuristic rule of setting;
Step 14 is scheduled the task in the observation mission set after the integration, according to sequence with determination one by one
Whether it is added toIn, it can be added without task again in the observation mission set after the integrationIn,
Step 15 exports the operation plan in subsequent time period T
In C*The dispatching algorithm of the readjustment degree moment point of driving is as follows:
Input:
- in this period of time T and it is later than C*Drive the operation plan of scheduling instance point t;
- arrived before scheduling instance point t and unscheduled emergent observation mission set;
Output:
- the operation plan revised in time t,
It is as follows:
Step 21 is in time t to next T- according to observation time window and drives in this time interval of scheduling instance point
Condition, from set of tasksIt is middle to choose observation mission of meeting an urgent need, generate new set of tasks
Step 22 is right according to the heuristic rule of settingIn emergent observation mission be ranked up;
Step 23 is chosen one by one according to new Mission Event SequenceIn emergent observation mission and rightIt is revised, untilIn can be added without emergent observation mission againIn,
Step 24 exports the operation plan revised
Preferably, a kind of intelligent remote sensing satellite stratification distributed freedom cotasking planning system includes structuring
Neural network module, the structural neural networks module carry out imaging task schedulability using structural neural networks pre-
It surveys, wherein the structural neural networks module is built by causality theories, all connection relations between each node
It is based on the causality structure of real system.
Preferably, the intelligent remote sensing satellite stratification distributed freedom cotasking planning system includes structuring god
Through network module, the structural neural networks module carries out imaging task schedulability using structural neural networks pre-
It surveys, wherein the structural neural networks module is built by causality theories, all connection relations between each node
It is based on the causality structure of real system.
The present invention also provides a kind of intelligent remote sensing satellite stratification distributed freedom cotasking planning systems comprising knot
Structure neural network module, learning type intellectual optimization module, constraint reasoning module;The structural neural networks module is using knot
Structure neural network predicts imaging task schedulability;The learning type intellectual optimization module is excellent using learning type intellectual
Change method carries out dynamic dispatchings to the more resources of multitask under higher dimensional space comprising learning-oriented genetic algorithm module and learning-oriented
Ant group algorithm module, the learning-oriented genetic algorithm module to the more resources of multitask roll by learning-oriented genetic algorithm divides
Match, the learning-oriented ant group algorithm module carries out rolling scheduling by learning-oriented ant group algorithm to the more earth stations of multitask;It is described
Constraint reasoning module plans single star from main task with constraint reasoning by intelligent optimization;The autonomous cotasking planning
Task is carried out pool distribution, and carry out distributed treatment by system by the distributed autonomous cotasking system of stratification.
Preferred scheme, the structural neural networks module are built by causality theories, institute between each node
Some connection relations are based on the causality structure of real system.
Further preferred scheme, the learning type intellectual optimization module are mutually tied using intelligent optimization model with knowledge model
The method of conjunction carries out integrated moulding;The intelligent optimization model treats the feas ible space of optimization problem according to " neighborhood search " strategy
It scans for, the knowledge model is useful knowledge to be excavated from the optimization process of early period, and instruct using obtained knowledge
The follow-up optimization process of intelligent optimization method.
Scheme still more preferably, the constraint reasoning module include logical constraint reasoning, Temporal resoning and resource about
Beam reasoning.
Scheme still further preferably, the logical constraint reasoning use condition triggering mode, are generated newly according to condition
Activity is simultaneously inserted into;The Temporal resoning uses the Trail consistency check and constraint propagation technique of Temporal Constraint Network, makes the time
Codomain is reduced and time-constrain meets;The resource constraint reasoning is built upon on the basis of time network, with resource time net
Network describes problem, and the distribution of computing resource consumption level finds defect according to distribution, and is based on mechanisms of defect management, and adjustment is lived
Constraint between dynamic.
The autonomous cotasking system include star task dispatcher more than one and it is multiple be independently distributed and with more stars
The connected single star task dispatcher of task dispatcher.
The task coordination method of the autonomous cotasking system is:More star task dispatchers receive new task, and to new
Task carries out task restriction parsing;Then task is assigned on each single star task dispatcher by allocation algorithm, and will appointed
Business is processed into the Meta task of single star task dispatcher Direct Recognition;Dispatching algorithm is finally executed by single star task dispatcher, is generated
Respectively the observation program of observation resource is to corresponding satellite.
It is connected for bidirectional information between one more star task dispatchers and multiple single star task dispatchers, each list star is appointed
Single star scheduling result of business scheduler feeds back to more star task dispatchers, and unfinished task is by more star task dispatchers according to it
The state of his satellite carries out sub-distribution again.
By using above technical scheme, intelligent remote sensing satellite stratification distributed freedom cotasking of the invention planning
Compared with prior art, advantage is system:
1, the present invention uses stratification distributed freedom cotasking Planning Model, the coordinator of master control rank and multiple lists
By feedback redistribution mechanism several times between star task dispatcher, the rationalization of allocation plan is promoted, to promote
The use of resource is more efficient.
2, layering mission planning carries out challenge to plan as a whole distribution, distributed treatment in advance, substantially reduces asking for problem
Complexity is solved, and uses task schedulability prediction technique, the linking reasonably established between two Hierarchical Decision Making variables is closed
System, task schedulability prediction can in the top-level task pre-planning stage, pre-estimate each dispatching platforms of lower layer as a result, in this, as
The foundation of task distribution avoids the blindness of prerequisite task distribution caused by the hysteresis quality of later stage scheduling result feedback.
3, the more resource dynamics of multitask that the present invention has successfully tackled key problems under the prediction of imaging task schedulability, higher dimensional space are adjusted
Key scientific problems and the corresponding technical barriers such as degree, fining planning and scheduling algorithm design, make the autonomous cotasking of more stars
Planning technology can be preferably applied for actual national defense construction field.
4, the present invention predicts imaging task schedulability using structural neural networks, structural neural networks mould
All connection relations are all based on what the causality of real real system constructed between each node of type, have stronger
Model parameter interpretability, efficiently solves various defects existing for conventional feed forward neural network model, and such as model is non-structural
Change, convergence rate is slow, neuron number is difficult determining and Local Minimum etc..
5, multiple by the way that complicated dynamic scheduling problem is decomposed into using the more resource dynamic rolling distribution mechanisms of multitask
Simple static scheduling subproblem, then the optimization solution of subproblem is combined, it is big in this way to replace the optimal solution of former problem
The difficulty of former problem solving is reduced greatly.
Description of the drawings
Fig. 1 is the autonomous cotasking planning system frame diagram of the intelligent remote sensing satellite network of the present invention;
Fig. 2 is the overall plan figure of the autonomous cotasking planning system of the intelligent remote sensing satellite network of the present invention;
Fig. 3 is the conceptual scheme of the learning type intellectual optimization method of the present invention;
Fig. 4 is the stratification distributed collaboration mission planning frame diagram of the present invention;
Fig. 5 is the system assumption diagram of the distributed autonomous cotasking system of stratification of the present invention.
Specific implementation mode
In order to make the objectives, technical solutions and advantages of the present invention clearer, with reference to specific example, to the present invention
It is further described.It should be understood that these descriptions are merely illustrative, and it is not intended to limit the scope of the present invention.In addition,
In illustrating below, descriptions of well-known structures and technologies are omitted, so as not to unnecessarily obscure the concept of the present invention.
The intelligent remote sensing satellite stratification distributed freedom cotasking planning system of the present invention includes more star task coordinates
Task in set of tasks to be allocated is distributed to more intelligence having under its command by scheduler on device and star, more star task dispatchers
Satellite and the Meta task that task is processed into scheduler Direct Recognition on star, each intelligent satellite is using scheduler on its star to being divided
The new task matched and existing task carry out United Dispatching, wherein more star task dispatchers are before carrying out task distribution, in advance
First estimate the scheduling result of scheduler on related star, and in this, as the foundation of task distribution, later stage scheduling result is avoided to feed back
Hysteresis quality.
It pre-estimates and any method appropriate or strategy may be used.A kind of optional method be according to task saturation degree or
Resources idle degree is estimated on star.For example, on star residue 20% resource, and the newly increasing of the task is estimated occupies 15%
Resource, then success can be dispatched by estimating.Another optional method is to be judged according to priority of task grade.For example, on star
The resource of residue 20%, and newly increasing for task is expected to occupy 25% resource, still, the priority level of waiting task on star
Less than new distribution task, then success can be dispatched by also estimating.As for being possible to be replaced of the task, then coordinator is retracted into again
It is allocated.Advantageously, carrying out above-mentioned pre-estimating or predicting using neural network.Thus, it is possible to have higher
Forecasting accuracy and prediction fineness.
Set of tasks in rolling window is distributed to more intelligent satellites having under its command, each intelligence by more star task dispatchers
Energy satellite is scheduled assigned new task and existing task using scheduler on its star, in the starting of current scrolling window
Moment, more star task dispatchers are updated mission bit stream, delete having been completed in a upper rolling window for task and
The task that the initial time is carrying out, and by unappropriated task in a upper rolling window and in a upper rolling window
The new task of interior arrival is combined into the set of tasks in current scrolling window, and more star task dispatchers are by the set of tasks
It is allocated to the more intelligent satellites, wherein the initial time of rolling window, a side are determined based on mixing triggering pattern
Face rolls distribution every the triggering of period, which is constant or dynamic change according to preset rules;Separately
On the one hand, there is the event for making system mode change or triggering rolling distribution when by manual intervention.
Preferably, on the one hand, the period is arranged according to the observing and controlling period;On the other hand, described to make system mode
The event of change includes:Emergent observation mission is received, and the unallocated emergent observation mission accumulated reaches five either institutes
State the 5% of the intelligent satellite number that more star task dispatchers have under its command.
In one alternate embodiment, more star task dispatchers include that earth station and geostationary orbit communication are defended
Star, within the observing and controlling period, the earth station carries out task distribution;Except the observing and controlling period, the geostationary orbit communication
Satellite carries out task distribution, and the emergent observation mission is generated by the intelligent satellite, wherein the geostationary orbit is logical
Satellite is believed only to the intelligent satellite with communications loop carries out task distribution therewith when being allocated, at this point, more stars are appointed
The intelligent satellite number that business coordinator has under its command refers to the intelligent satellite for having communications loop with the geostationary orbit telecommunication satellite
Quantity.
Preferably, the task scheduling strategy of scheduler is as follows on the star of each intelligent satellite:
(1) next week is generated using the complete rescheduling strategy in gradual method in the scheduling instance point of T- drivings
New task plan in phase time interval, the scheduling instance point of T- drivings are determined specifically according to given time interval T
Scheduling time point lT, 0≤l≤L, LT≤H < (L+1) T often reach a scheduling time point lT, then calculate and generate latter scheduling
The task scheduling in section [lT, (l+1) T], wherein l are positive integer, and T is given time interval, and L is maximum T- driving scheduling time
Number, H are total activation section,
(2) in C*The readjustment degree moment point of driving works as satellite using the operation plan correcting strategy in revision formula method
When operating in given scheduling interval, if at a time t (0 < t < H), the emergent observation mission cumulant C on startIt is super
Cross given threshold value C*When, then execute the calculating of readjustment degree, wherein threshold value C*For meet an urgent need observation mission critical integral number,
In addition to above two scheduling instance point, scheduler is not scheduled in any other moment point on star.
Advantageously, in the case where the rolling window of more star task dispatchers is the constant period, the tune of T- drivings
The period for spending moment point is equal to the constant period.In the time that the rolling window of more star task dispatchers is rule variation
In the case of section, the period of the scheduling instance point of T- drivings is equal to or less than the minimum length of the period.
In another alternative embodiment, whenever receiving the newly assigned task of more star task dispatchers, scheduler on star
Just task is once dispatched, and feedback scheduling result.
Specifically, as follows in the dispatching algorithm of the scheduling instance point of T- drivings:
Input:
The emergent observation mission set that arrived and be not scheduled before T- drives scheduling instance point;
The routine observation set of tasks for having received and not being scheduled before T- drives scheduling instance point;
Output:
-- the operation plan in subsequent time period T;
It is as follows:
Step 11 respectively fromWithWhether middle access time window falls into the routine observation in next period of time T
Task and emergent observation mission generate the routine observation set of tasks for waiting for that scheduling solvesWith emergent observation mission set
Step 12 willWithIt is integrated into an observation mission set;
Step 13 is ranked up the task in the observation mission set after integration according to the heuristic rule of setting;
Step 14 is scheduled the task in the observation mission set after the integration, according to sequence with determination one by one
Whether it is added toIn, it can be added without task again in the observation mission set after the integrationIn,
Step 15 exports the operation plan in subsequent time period T
In C*The dispatching algorithm of the readjustment degree moment point of driving is as follows:
Input:
- in this period of time T and it is later than C*Drive the operation plan of scheduling instance point t;
- arrived before scheduling instance point t and unscheduled emergent observation mission set;
Output:
- the operation plan revised in time t,
It is as follows:
Step 21 is in time t to next T- according to observation time window and drives in this time interval of scheduling instance point
Condition, from set of tasksIt is middle to choose observation mission of meeting an urgent need, generate new set of tasks
Step 22 is right according to the heuristic rule of settingIn emergent observation mission be ranked up;
Step 23 is chosen one by one according to new Mission Event SequenceIn emergent observation mission and rightIt is revised, untilIn can be added without emergent observation mission againIn,
Step 24 exports the operation plan revised
Preferably, a kind of intelligent remote sensing satellite stratification distributed freedom cotasking planning system includes structuring
Neural network module, the structural neural networks module carry out imaging task schedulability using structural neural networks pre-
It surveys, wherein the structural neural networks module is built by causality theories, all connection relations between each node
It is based on the causality structure of real system.
Preferably, the intelligent remote sensing satellite stratification distributed freedom cotasking planning system includes structuring god
Through network module, the structural neural networks module carries out imaging task schedulability using structural neural networks pre-
It surveys, wherein the structural neural networks module is built by causality theories, all connection relations between each node
It is based on the causality structure of real system.
The present invention is as shown in Figure 1, it plans framework, based on structuring nerve around stratification distributed freedom cotasking
The imaging task schedulability prediction of network, the more resources of the multitask based on learning-oriented genetic algorithm roll distribution, based on intelligence
The more earth station's rolling schedulings of the autonomous mission planning of single star, the multitask based on learning-oriented ant group algorithm etc. of optimization and constraint reasoning
Five main researchs, the more resource dynamics of multitask that emphasis is tackled key problems under the prediction of imaging task schedulability, higher dimensional space are adjusted
Key scientific problems and the corresponding technical methods such as degree, fining planning and scheduling algorithm design, make the autonomous cotasking of more stars
Planning technology can be preferably applied for actual national defense construction field.
The present invention has mainly used the researchs sides such as structural neural networks, learning type intellectual optimization method and constraint reasoning
Method, overall plan are as shown in Figure 2.
(1) structural neural networks.The limitation in view of standard neural network with " black-box model ", present invention employs
Structural neural networks predict imaging task schedulability.Structural neural networks structure based on causality theories
It builds, all connection relations are all based on what the causality of real real system constructed between each node, have relatively strong
Model parameter interpretability (each parameter have actual interpretable meaning).Structural neural networks model effectively solves
Various defects existing for conventional feed forward neural network model, such as model is unstructured, convergence rate is slow, neuron number very
Difficult determination and Local Minimum etc..
, by construction to mission planning result sample set characteristic value and extraction, design is a kind of adjustable for imaging task for it
The structural neural networks model of degree property prediction;Primary study how a large amount of operations based on the accumulation of satellite task planning system
Data dynamic adjusts the structural model and relevant parameter of structural neural networks.Structural neural networks model is in learning process
The Nonlinear Mapping relationship between task characteristic value and satellite capacity can be established, so as to the schedulability to imaging task
Prediction.
(2) learning type intellectual optimization method.In doctor's conceptual phase, inventor is based on developing to construct with study mechanism asking
Learning type intellectual optimization method of the solution towards complicated optimum problem:It is combined using intelligent optimization model and knowledge model integrated
Modeling approach, the feas ible space that intelligent optimization model treats optimization problem according to " neighborhood search " strategy scan for;Knowledge mould
Type excavates some useful knowledge from the optimization process of early period, then instructs intelligent optimization method using obtained knowledge
Follow-up optimization process.The present invention solves the more resources of multitask using learning-oriented genetic algorithm and learning-oriented ant group algorithm and rolls respectively
The complicated optimum problems such as distribution, the more earth station's rolling schedulings of multitask.
The present invention is directed to the extensive characteristic that the autonomous cotasking planning problem of intelligent satellite network shows, and synthesis is examined
Worry task arrival time and observing and controlling time window dynamic divide dispatching cycle, and shorter week is pressed using heuristic rule and prediction mechanism
Phase rolls adjustment initial planning plan, forms the dynamic rolling programming dispatching technology of task-driven, realizes to dynamic environment and task
The quick response of demand.By Genetic Optimization model and knowledge model it is integrated after form learning-oriented genetic algorithm, can by tens thousand of at
As task is dynamically assigned to hundreds of remote sensing satellites, provided for other similar higher-dimension Solving Assignment Problems a kind of beneficial to borrowing
Mirror.Learning-oriented ant group algorithm is formed after ant group optimization model and knowledge model are integrated, can dynamically be adjusted in hundreds of earth stations
Ten thousand number biography tasks of the number of degrees provide effective technical support for other similar higher-dimension Scheduling Problems.Solving big rule
In the learning type intellectual optimization method of mould complicated optimum problem, innovation algorithm and its core operation proposed by the present invention such as Fig. 3 institutes
Show.
The more resources of multitask based on learning-oriented genetic algorithm roll distribution technique, are based on roll stablized loop principle structure
The scroll mechanism for having built the more Resource dynamic allocations of multitask converts complicated dynamic allocation problem to and rolls newer static allocation
Problem;If definition can embody Ganlei's knowledge of the more resource allocation problem substantive characteristics of multitask, structure can effectively manage these and know
The knowledge model of knowledge;Genetic Optimization model based on the more resource allocation problem feasible program structures of genetic Algorithm Design multitask;
Integrated and interaction mechanism of the primary study between Genetic Optimization model and knowledge model, ultimately form Genetic Optimization model and
The learning-oriented genetic algorithm that knowledge model efficiently integrates.
The more earth station's rolling scheduling technologies of multitask based on learning-oriented ant group algorithm are based on roll stablized loop principle
The scroll mechanism for building the more earth station's dynamic dispatchings of multitask converts complicated dynamic scheduling problem to and rolls newer static tune
Degree problem;On the basis of arranging and summarizing the expertise, user preference and posterior infromation in earth station's scheduling of resource field, structure
The knowledge model for ancillary terrestrial station scheduling of resource is built;Based on the more earth station's scheduling problems of ant colony algorithm for optimization design multitask
The ant group optimization model of feasible program structure;Integrated between ant group optimization model and knowledge model of primary study and interact machine
System ultimately forms the learning-oriented ant group algorithm for efficiently integrating ant group optimization model and knowledge model.
(3) constraint reasoning technology.Constraint reasoning includes mainly three parts:Logical constraint reasoning, Temporal resoning and resource are about
Beam reasoning.Reasoning from logic mainly uses condition to trigger, and generates New activity according to condition and is inserted into.Temporal resoning mainly uses the time
The Trail consistency check and constraint propagation technique of constraint network realize the satisfaction of the reduction and time-constrain of time codomain.Resource
Reasoning is built upon on the basis of time network, and problem is described with resource time network, since activity changes resource with relative mode
State, it is therefore desirable to which the distribution of computing resource consumption level finds defect according to distribution, is based on mechanisms of defect management, and adjustment is lived
It is dynamic to be constrained between activity.
The autonomous mission planning technology of single star based on intelligent optimization and constraint reasoning, by domain model, time and resource
The portions modulars such as constraint reasoning and problem model, integrated attitude control model, deposit the space flight such as model and antenna model at battery model admittedly
Domain model constructs the Integrated Planning and Scheduling Framework having the scalability with versatility;Structure is based on intelligent optimization and constraint
The autonomous mission planning technology of single star of reasoning:Intelligent optimization module can be chosen for union variable with task and moving machine and carry out part
Search, constraint reasoning module carries out processing to logical relation, time and the resource constraint in Task-decomposing activity diagram and conflict disappears
Solution.Consider to generate using the relevant heuristic information of satellite fields and the reasoning of user preference guided constraint and plan, in terms of less
Calculation amount produces better mission planning result.
The technology path and laboratory facilities of each research contents of the invention are provided in turn below.
1, stratification distributed freedom cotasking plans framework
Dual layer resist is theoretical and model has unique fit in terms of coping with the decision optimization problem with multi-level characteristic
Ying Xing, the autonomous cotasking planning problem of more stars being also very suitable under distributed collaboration mechanism.Distributed collaboration it is emphasised that
Pass through the information exchange of top layer coordination unit between each subproblem.More autonomous cotasking planning of star are asked under distributed collaboration mechanism
Topic is suitble to be described using Bilevel Programming Problem mathematical model, and correlation modeling solution technique can be used for reference:More stars are independently assisted
It is divided into the multi-platform multitask coordinated distribution of top layer with planning process and the contexture by self two of bottom list platform be combined with each other, closely
The decision process (such as Fig. 4) of connection.Fig. 4 at the middle and upper levels be multi-platform multitask dynamically distribute, by the assigning process will according to appoint
Business feature and resource characteristics distribute task to each observation resource.
Stratification distributed freedom cotasking Planning Model, one is increased on the basis of centralized cotasking is planned
More star task dispatchers of a master control rank, and multi satellites joint scheduler is eliminated, the scheduling of every satellite is still made
With its dedicated single star task scheduling, as shown in figure 5, more star task dispatcher management and control ranks are higher, more star task dispatchers pair
Task carry out task restriction parsing, according to mission requirements and have under its command observation resource state, by specific distribution algorithm will appoint
Business is assigned to each observation resource up, and task is processed into the Meta task of single star scheduler Direct Recognition, then by single star tune
It spends device and executes the observation program that dispatching algorithm generates respectively observation resource.Each list star scheduler can be anti-to more star task dispatchers
The single star scheduling result of feedback, for unfinished task, more star task dispatchers can carry out again according to the state of other satellites
Distribution, by feedback redistribution mechanism several times, can promote the rationalization of allocation plan, to promote resource to use more
Increase effect.When new task arrival is less, more star task dispatchers can be according to task feature and existing single star task execution
The task is distributed to some or a few satellites by scheme, to trigger the scheduling flow of this several satellites, for not there is new point
Single star with task continues to execute existing program, to realize the asynchronous management and control of more stars, enhances the flexible of observation resource management
Property.This pattern avoids the complexity of more stars scheduling unified Modelings to a certain extent, by the carry out layered shaping of scheduling problem,
The reusability for enhancing system also enhances system extension malleability, if interim increase or reduce collaboration satellite temporarily, only needs
It to modify at more star task dispatchers.
It is layered mission planning to carry out challenge to plan as a whole distribution, distributed treatment in advance, greatly reduces asking for problem
Complexity is solved, however can the joining relation that rationally establish between two Hierarchical Decision Making variables be then to determine that layering mission planning has
The key of effect property.The present invention solves this linking problem by studying rational task schedulability prediction technique.Task can
Scheduling property prediction effect be exactly in the top-level task pre-planning stage, pre-estimate each dispatching platforms of lower layer as a result, in this, as
The foundation of task distribution, avoids the blindness that prerequisite task is distributed caused by the hysteresis quality that later stage scheduling result is fed back
Property.For this purpose, the structure by carrying out sample set characteristic value to the actual, historical data that early period, high score system administration scheduling process accumulated
It makes and extracts, and establish the agent model of scheduling of resource based on integrated BP neural network, to be provided to scheduling by the model
As a result prediction.When actual schedule result online feedback, model can be updated.
The above specific embodiments are only exemplary, is to preferably make skilled artisans appreciate that originally
Patent, be not to be construed as include to this patent range limitation;As long as appointing according to made by spirit disclosed in this patent
How with change or modification, the range that this patent includes is each fallen within.
Claims (7)
1. a kind of intelligence remote sensing satellite stratification distributed freedom cotasking planning system, which is characterized in that appoint including more stars
Be engaged in scheduler on coordinator and star, more star task dispatchers the task in set of tasks to be allocated is distributed to have under its command it is more
Intelligent satellite and the Meta task that task is processed into scheduler Direct Recognition on star, each intelligent satellite utilize scheduler on its star
United Dispatching is carried out to assigned new task and existing task, wherein more star task dispatchers are carrying out task distribution
Before, the scheduling result of scheduler on related star is pre-estimated, and in this, as the foundation of task distribution.
2. intelligence remote sensing satellite stratification distributed freedom cotasking planning system according to claim 1, feature
It is, the set of tasks in rolling window is distributed to more intelligent satellites having under its command, each intelligence by more star task dispatchers
Satellite is scheduled assigned new task and existing task using scheduler on its star, in the starting of current scrolling window
It carves, more star task dispatchers are updated mission bit stream, delete having been completed in a upper rolling window for task and in institute
The task that initial time is carrying out is stated, and by unappropriated task in a upper rolling window and in a upper rolling window
The new task of arrival is combined into the set of tasks in current scrolling window, and more star task dispatchers by the set of tasks to
The more intelligent satellites are allocated, wherein determine the initial time of rolling window, a side based on mixing triggering pattern
Face rolls distribution every the triggering of period, which is constant or dynamic change according to preset rules;Separately
On the one hand, there is the event for making system mode change or triggering rolling distribution when by manual intervention.
3. intelligence remote sensing satellite stratification distributed freedom cotasking planning system according to claim 2, feature
It is, on the one hand, the period is arranged according to the observing and controlling period;On the other hand, described to make the event that system mode changes
Including:Emergent observation mission is received, and the unallocated emergent observation mission accumulated reaches five either more star tasks
The 5% of the intelligent satellite number that coordinator has under its command.
4. the more resources of a kind of multitask based on learning-oriented genetic algorithm according to claim 3 roll distribution method,
It being characterized in that, more star task dispatchers include earth station and geostationary orbit telecommunication satellite, within the observing and controlling period, institute
It states earth station and carries out task distribution;Except the observing and controlling period, the geostationary orbit telecommunication satellite carries out task distribution, and institute
It states emergent observation mission to be generated by the intelligent satellite, wherein the geostationary orbit telecommunication satellite is only to being allocated
When therewith with communications loop intelligent satellite carry out task distribution, at this point, the intelligence that more star task dispatchers have under its command is defended
Star number refers to the quantity for the intelligent satellite for having communications loop with the geostationary orbit telecommunication satellite.
5. the intelligent remote sensing satellite stratification distributed freedom cotasking planning system according to any one of claim 1-4
System, which is characterized in that the task scheduling strategy of scheduler is as follows on the star of each intelligent satellite:
(1) in the scheduling instance point of T- drivings, using the complete rescheduling strategy in gradual method, when generating next cycle
Between new task plan in section, the scheduling instance point of T- drivings is to determine specific scheduling according to given time interval T
Time point lT, 0≤l≤L, LT≤H < (L+1) T often reaches a scheduling time point lT, then calculates and generate latter scheduling interval
The task scheduling of [lT, (l+1) T], wherein l are positive integer, and T is given time interval, and L is that maximum T- drives scheduling times, H
For total activation section,
(2) in C*The readjustment degree moment point of driving, using the operation plan correcting strategy in revision formula method, when satellite transit exists
When in given scheduling interval, if at a time t (0 < t < H), the emergent observation mission cumulant C on startMore than given
Threshold value C*When, then execute the calculating of readjustment degree, wherein threshold value C*For meet an urgent need observation mission critical integral number,
In addition to above two scheduling instance point, it is not scheduled in any other moment point,
It is as follows in the dispatching algorithm of the scheduling instance point of T- drivings:
Input:
The emergent observation mission set that arrived and be not scheduled before T- drives scheduling instance point;
The routine observation set of tasks for having received and not being scheduled before T- drives scheduling instance point;
Output:
-- the operation plan in subsequent time period T;
It is as follows:
Step 11 respectively fromWithWhether middle access time window falls into the routine observation task in next period of time T
With emergent observation mission, the routine observation set of tasks for waiting for that scheduling solves is generatedWith emergent observation mission set
Step 12 willWithIt is integrated into an observation mission set;
Step 13 is ranked up the task in the observation mission set after integration according to the heuristic rule of setting;
Step 14 is scheduled the task in the observation mission set after the integration, according to sequence to determine whether one by one
It is added toIn, it can be added without task again in the observation mission set after the integrationIn,
Step 15 exports the operation plan in subsequent time period T
In C*The dispatching algorithm of the readjustment degree moment point of driving is as follows:
Input:
- in this period of time T and it is later than C*Drive the operation plan of scheduling instance point t;
- arrived before scheduling instance point t and unscheduled emergent observation mission set;
Output:
- the operation plan revised in time t,
It is as follows:
Step 21 is in the item in time t to next T- driving scheduling instance point this time interval according to observation time window
Part, from set of tasksIt is middle to choose observation mission of meeting an urgent need, generate new set of tasks
Step 22 is right according to the heuristic rule of settingIn emergent observation mission be ranked up;
Step 23 is chosen one by one according to new Mission Event SequenceIn emergent observation mission and rightIt is revised, until
In can be added without emergent observation mission againIn,
Step 24 exports the operation plan revised
6. a kind of intelligent remote sensing satellite stratification distributed freedom cotasking rule according to any one of claim 1-4
The system of drawing, which is characterized in that including structural neural networks module, the structural neural networks module is using structuring nerve
Network predicts imaging task schedulability, wherein the structural neural networks module is by causality theories
It builds, all connection relations are based on the causality structure of real system between each node.
7. a kind of intelligent remote sensing satellite stratification distributed freedom cotasking planning system according to claim 6,
Be characterized in that, including structural neural networks module, the structural neural networks module using structural neural networks at
As task schedulability is predicted, wherein the structural neural networks module is built by causality theories, each to save
All connection relations are based on the causality structure of real system between point.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711433054.7A CN108335012A (en) | 2017-12-26 | 2017-12-26 | A kind of intelligence remote sensing satellite stratification distributed freedom cotasking planning system |
PCT/CN2018/080422 WO2019127948A1 (en) | 2017-12-26 | 2018-03-26 | Hierarchical distributed autonomous collaborative task planning system for intelligent remote sensing satellite |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711433054.7A CN108335012A (en) | 2017-12-26 | 2017-12-26 | A kind of intelligence remote sensing satellite stratification distributed freedom cotasking planning system |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108335012A true CN108335012A (en) | 2018-07-27 |
Family
ID=62924551
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711433054.7A Pending CN108335012A (en) | 2017-12-26 | 2017-12-26 | A kind of intelligence remote sensing satellite stratification distributed freedom cotasking planning system |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN108335012A (en) |
WO (1) | WO2019127948A1 (en) |
Cited By (26)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109087023A (en) * | 2018-08-23 | 2018-12-25 | 中南大学 | A kind of more stars observation layering dispatching method and system based on divide-and-conquer strategy |
CN109214564A (en) * | 2018-08-30 | 2019-01-15 | 北京控制工程研究所 | A kind of remote sensing of the earth satellite Autonomous mission planning method with a variety of Planning Models |
CN109299861A (en) * | 2018-09-04 | 2019-02-01 | 中国人民解放军国防科技大学 | On-satellite scheduling method of agile satellite |
CN109470266A (en) * | 2018-11-02 | 2019-03-15 | 佛山科学技术学院 | A kind of star sensor Gyro method for determining posture handling multiplicative noise |
CN109739638A (en) * | 2018-12-06 | 2019-05-10 | 北京航空航天大学 | A kind of EDF schedulability determination method and device based on deep learning |
CN109767128A (en) * | 2019-01-15 | 2019-05-17 | 中国人民解放军国防科技大学 | imaging satellite autonomous task planning method based on machine learning |
CN109766181A (en) * | 2018-12-06 | 2019-05-17 | 北京航空航天大学 | A kind of RMS schedulability determination method and device based on deep learning |
CN111046320A (en) * | 2019-12-12 | 2020-04-21 | 南京航空航天大学 | Dynamic allocation method for remote sensing satellite mission planning energy |
CN111311074A (en) * | 2020-01-20 | 2020-06-19 | 中国人民解放军国防科技大学 | Multi-satellite distributed cooperative rescheduling method facing emergency tasks |
CN111414998A (en) * | 2020-04-21 | 2020-07-14 | 中国人民解放军国防科技大学 | Remote sensing satellite resource scheduling algorithm selection method based on deep neural network |
CN111914483A (en) * | 2020-08-03 | 2020-11-10 | 北京理工大学 | Multi-spacecraft autonomous task planning time constraint distributed processing method |
CN111930351A (en) * | 2020-08-07 | 2020-11-13 | 上海卫星工程研究所 | Task planning software implementation method based on real-time operating system |
CN111950874A (en) * | 2020-07-30 | 2020-11-17 | 上海卫星工程研究所 | Satellite autonomous task planning satellite-borne knowledge base design method and system |
CN111950880A (en) * | 2020-07-31 | 2020-11-17 | 上海卫星工程研究所 | Multi-satellite cooperative task description and on-satellite decomposition processing method and system |
CN112183929A (en) * | 2018-11-07 | 2021-01-05 | 长沙天仪空间科技研究院有限公司 | Imaging system of remote sensing satellite |
CN112232594A (en) * | 2020-11-06 | 2021-01-15 | 中国人民解放军国防科技大学 | Multi-satellite emergency observation task planning method and device and computer equipment |
CN112612617A (en) * | 2020-12-30 | 2021-04-06 | 东方红卫星移动通信有限公司 | Satellite telemetry data processing method and system and constellation state monitoring platform |
CN112633458A (en) * | 2020-12-09 | 2021-04-09 | 西安电子科技大学 | Matching method and device for solving scheduling problem of satellite measurement, operation and control resource planning |
CN112734107A (en) * | 2021-01-08 | 2021-04-30 | 浙江大学 | Product collaborative design task data decomposition and distribution method based on error inverse propagation |
CN113010840A (en) * | 2020-08-04 | 2021-06-22 | 南京航空航天大学 | Measurement task planning method based on task similarity |
CN113128828A (en) * | 2021-03-05 | 2021-07-16 | 中国科学院国家空间科学中心 | Satellite observation distributed online planning method based on multi-agent reinforcement learning |
CN113220425A (en) * | 2021-03-17 | 2021-08-06 | 中国人民解放军战略支援部队航天工程大学 | Distributed reconfigurable satellite system organization method based on mosaic splicing |
CN113330470A (en) * | 2019-01-25 | 2021-08-31 | 西门子股份公司 | Autonomous coordination of devices in an industrial environment |
CN113986830A (en) * | 2021-11-11 | 2022-01-28 | 西安交通大学 | Distributed CT-oriented cloud data management and task scheduling method and system |
CN115510250A (en) * | 2022-11-10 | 2022-12-23 | 成都国恒空间技术工程股份有限公司 | Knowledge block chain-based aerospace information collaborative acquisition method and system |
CN116151039A (en) * | 2023-04-19 | 2023-05-23 | 中国西安卫星测控中心 | Distributed collaborative scheduling method and system based on random access task |
Families Citing this family (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111176807B (en) * | 2019-12-09 | 2023-12-12 | 中国空间技术研究院 | Multi-star collaborative task planning method |
CN111444007B (en) * | 2020-06-16 | 2020-11-17 | 南京理工大学 | Remote sensing big data automatic processing method based on cloud computing |
CN113537782B (en) * | 2021-07-19 | 2023-08-18 | 福州大学 | Multi-satellite situation awareness system distributed task planning method based on contract network |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20140039963A1 (en) * | 2012-08-03 | 2014-02-06 | Skybox Imaging, Inc. | Satellite scheduling system |
CN104050324A (en) * | 2014-06-23 | 2014-09-17 | 中国人民解放军国防科学技术大学 | Mathematical model construction method and solving method for single-star task planning problem |
CN106845793A (en) * | 2016-12-27 | 2017-06-13 | 中国电子科技集团公司第五十四研究所 | Roller remote sensing satellite dynamic task planing method based on observing and controlling opportunity |
-
2017
- 2017-12-26 CN CN201711433054.7A patent/CN108335012A/en active Pending
-
2018
- 2018-03-26 WO PCT/CN2018/080422 patent/WO2019127948A1/en active Application Filing
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20140039963A1 (en) * | 2012-08-03 | 2014-02-06 | Skybox Imaging, Inc. | Satellite scheduling system |
CN104050324A (en) * | 2014-06-23 | 2014-09-17 | 中国人民解放军国防科学技术大学 | Mathematical model construction method and solving method for single-star task planning problem |
CN106845793A (en) * | 2016-12-27 | 2017-06-13 | 中国电子科技集团公司第五十四研究所 | Roller remote sensing satellite dynamic task planing method based on observing and controlling opportunity |
Non-Patent Citations (4)
Title |
---|
GUOLIANG LI等: "A hybrid Online Scheduling mechanism with revision and progressive techniques for Autonomous Earth observation Satellite", 《ACTA ASTRONAUTICA》 * |
义余江 等: "面向动态环境的成像卫星自主任务规划方法", 《中国系统工程学会第九次全国会员代表大会暨第18届学术年会》 * |
白保存 等: "多星任务规划系统调度结构", 《计算机工程》 * |
邢立宁: "面向新型遥感卫星的星上自主任务规划框架", 《第三届高分辨率对地观测学术年会优秀论文集》 * |
Cited By (41)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109087023B (en) * | 2018-08-23 | 2021-03-02 | 中南大学 | Multi-satellite observation layered scheduling method and system based on divide-and-conquer strategy |
CN109087023A (en) * | 2018-08-23 | 2018-12-25 | 中南大学 | A kind of more stars observation layering dispatching method and system based on divide-and-conquer strategy |
CN109214564B (en) * | 2018-08-30 | 2020-09-18 | 北京控制工程研究所 | Autonomous task planning method for ground remote sensing satellite with multiple planning modes |
CN109214564A (en) * | 2018-08-30 | 2019-01-15 | 北京控制工程研究所 | A kind of remote sensing of the earth satellite Autonomous mission planning method with a variety of Planning Models |
CN109299861A (en) * | 2018-09-04 | 2019-02-01 | 中国人民解放军国防科技大学 | On-satellite scheduling method of agile satellite |
CN109470266A (en) * | 2018-11-02 | 2019-03-15 | 佛山科学技术学院 | A kind of star sensor Gyro method for determining posture handling multiplicative noise |
CN112183929A (en) * | 2018-11-07 | 2021-01-05 | 长沙天仪空间科技研究院有限公司 | Imaging system of remote sensing satellite |
CN112183929B (en) * | 2018-11-07 | 2024-04-26 | 长沙天仪空间科技研究院有限公司 | Imaging system of remote sensing satellite |
CN109766181A (en) * | 2018-12-06 | 2019-05-17 | 北京航空航天大学 | A kind of RMS schedulability determination method and device based on deep learning |
CN109739638A (en) * | 2018-12-06 | 2019-05-10 | 北京航空航天大学 | A kind of EDF schedulability determination method and device based on deep learning |
CN109767128A (en) * | 2019-01-15 | 2019-05-17 | 中国人民解放军国防科技大学 | imaging satellite autonomous task planning method based on machine learning |
CN113330470A (en) * | 2019-01-25 | 2021-08-31 | 西门子股份公司 | Autonomous coordination of devices in an industrial environment |
CN111046320A (en) * | 2019-12-12 | 2020-04-21 | 南京航空航天大学 | Dynamic allocation method for remote sensing satellite mission planning energy |
CN111046320B (en) * | 2019-12-12 | 2023-10-13 | 南京航空航天大学 | Dynamic allocation method for remote sensing satellite mission planning energy |
CN111311074A (en) * | 2020-01-20 | 2020-06-19 | 中国人民解放军国防科技大学 | Multi-satellite distributed cooperative rescheduling method facing emergency tasks |
CN111311074B (en) * | 2020-01-20 | 2022-04-05 | 中国人民解放军国防科技大学 | Multi-satellite distributed cooperative rescheduling method facing emergency tasks |
CN111414998B (en) * | 2020-04-21 | 2024-01-19 | 中国人民解放军国防科技大学 | Remote sensing satellite resource scheduling algorithm selection method based on deep neural network |
CN111414998A (en) * | 2020-04-21 | 2020-07-14 | 中国人民解放军国防科技大学 | Remote sensing satellite resource scheduling algorithm selection method based on deep neural network |
CN111950874A (en) * | 2020-07-30 | 2020-11-17 | 上海卫星工程研究所 | Satellite autonomous task planning satellite-borne knowledge base design method and system |
CN111950874B (en) * | 2020-07-30 | 2023-12-12 | 上海卫星工程研究所 | Satellite autonomous mission planning satellite-borne knowledge base design method and system |
CN111950880A (en) * | 2020-07-31 | 2020-11-17 | 上海卫星工程研究所 | Multi-satellite cooperative task description and on-satellite decomposition processing method and system |
CN111950880B (en) * | 2020-07-31 | 2024-03-12 | 上海卫星工程研究所 | Multi-star collaborative task description and on-satellite decomposition processing method and system |
CN111914483B (en) * | 2020-08-03 | 2023-01-24 | 北京理工大学 | Multi-spacecraft autonomous task planning time constraint distributed processing method |
CN111914483A (en) * | 2020-08-03 | 2020-11-10 | 北京理工大学 | Multi-spacecraft autonomous task planning time constraint distributed processing method |
CN113010840A (en) * | 2020-08-04 | 2021-06-22 | 南京航空航天大学 | Measurement task planning method based on task similarity |
CN111930351B (en) * | 2020-08-07 | 2024-03-19 | 上海卫星工程研究所 | Task planning software implementation method based on real-time operating system |
CN111930351A (en) * | 2020-08-07 | 2020-11-13 | 上海卫星工程研究所 | Task planning software implementation method based on real-time operating system |
CN112232594A (en) * | 2020-11-06 | 2021-01-15 | 中国人民解放军国防科技大学 | Multi-satellite emergency observation task planning method and device and computer equipment |
CN112633458A (en) * | 2020-12-09 | 2021-04-09 | 西安电子科技大学 | Matching method and device for solving scheduling problem of satellite measurement, operation and control resource planning |
CN112612617A (en) * | 2020-12-30 | 2021-04-06 | 东方红卫星移动通信有限公司 | Satellite telemetry data processing method and system and constellation state monitoring platform |
CN112612617B (en) * | 2020-12-30 | 2023-06-20 | 东方红卫星移动通信有限公司 | Satellite telemetry data processing method and system and constellation state monitoring platform |
CN112734107A (en) * | 2021-01-08 | 2021-04-30 | 浙江大学 | Product collaborative design task data decomposition and distribution method based on error inverse propagation |
CN112734107B (en) * | 2021-01-08 | 2024-05-10 | 浙江大学 | Product collaborative design task data decomposition and distribution method based on error back propagation |
CN113128828B (en) * | 2021-03-05 | 2022-03-08 | 中国科学院国家空间科学中心 | Satellite observation distributed online planning method based on multi-agent reinforcement learning |
CN113128828A (en) * | 2021-03-05 | 2021-07-16 | 中国科学院国家空间科学中心 | Satellite observation distributed online planning method based on multi-agent reinforcement learning |
CN113220425A (en) * | 2021-03-17 | 2021-08-06 | 中国人民解放军战略支援部队航天工程大学 | Distributed reconfigurable satellite system organization method based on mosaic splicing |
CN113986830B (en) * | 2021-11-11 | 2024-02-23 | 西安交通大学 | Cloud data management and task scheduling method and system for distributed CT |
CN113986830A (en) * | 2021-11-11 | 2022-01-28 | 西安交通大学 | Distributed CT-oriented cloud data management and task scheduling method and system |
CN115510250B (en) * | 2022-11-10 | 2023-02-14 | 成都国恒空间技术工程股份有限公司 | Knowledge block chain-based aerospace information collaborative acquisition method and system |
CN115510250A (en) * | 2022-11-10 | 2022-12-23 | 成都国恒空间技术工程股份有限公司 | Knowledge block chain-based aerospace information collaborative acquisition method and system |
CN116151039A (en) * | 2023-04-19 | 2023-05-23 | 中国西安卫星测控中心 | Distributed collaborative scheduling method and system based on random access task |
Also Published As
Publication number | Publication date |
---|---|
WO2019127948A1 (en) | 2019-07-04 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108335012A (en) | A kind of intelligence remote sensing satellite stratification distributed freedom cotasking planning system | |
CN108256671A (en) | A kind of more resources of multitask based on learning-oriented genetic algorithm roll distribution method | |
WO2020181896A1 (en) | Multi-agent reinforcement learning scheduling method and system and electronic device | |
CN101237469B (en) | Method for optimizing multi-QoS grid workflow based on ant group algorithm | |
Qi et al. | Scalable parallel task scheduling for autonomous driving using multi-task deep reinforcement learning | |
CN108717572A (en) | Imaging task schedulability prediction technique based on structural neural networks | |
US5619695A (en) | Method and apparatus for scheduling resources | |
CN110427261A (en) | A kind of edge calculations method for allocating tasks based on the search of depth Monte Carlo tree | |
Miyashita | CAMPS: a constraint-based architecturefor multiagent planning and scheduling | |
Ghavamzadeh et al. | Learning to communicate and act using hierarchical reinforcement learning | |
Sauer et al. | Multi-site scheduling with fuzzy concepts | |
CN109714097B (en) | Cross-domain cooperative scheduling system for satellite resources | |
CN108304958A (en) | A kind of list star multi-load Complex Constraints mission planning method and system | |
CN109409773A (en) | A kind of earth observation resource dynamic programming method based on Contract Net Mechanism | |
CN111950870A (en) | Space-ground-based measurement and control data transmission resource integrated on-demand scheduling method and system | |
CN109343945A (en) | A kind of multitask dynamic allocation method based on contract net algorithm | |
WO2019127947A1 (en) | Intelligent optimization and constraint reasoning-based single-satellite autonomous task planning method | |
CN108171443A (en) | Consider the more star contingency tasks method and device for planning noted in instruction | |
Yang et al. | Onboard coordination and scheduling of multiple autonomous satellites in an uncertain environment | |
CN104077634A (en) | Active-reactive type dynamic project scheduling method based on multi-objective optimization | |
CN116893861A (en) | Multi-agent cooperative dependency task unloading method based on space-ground cooperative edge calculation | |
Ren et al. | Hierarchical reinforcement-learning for real-time scheduling of agile satellites | |
Lam et al. | DEVELOPMENT OF AN INTELLIGENT AGENT FOR AIRPORT GATE ASSIGNMENT. | |
Tharumarajah et al. | A behavior-based approach to scheduling in distributed manufacturing systems | |
Smith et al. | Continuous management of airlift and tanker resources: A constraint-based approach |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20180727 |
|
RJ01 | Rejection of invention patent application after publication |