CN117575371A

CN117575371A - Multi-star imaging task planning method and system based on man-machine interaction

Info

Publication number: CN117575371A
Application number: CN202410057937.6A
Authority: CN
Inventors: 胡敏; 杨学颖; 黄刚; 宋俊玲; 张锐; 李安迪; 齐晶; 黄飞耀
Original assignee: Peoples Liberation Army Strategic Support Force Aerospace Engineering University
Current assignee: Peoples Liberation Army Strategic Support Force Aerospace Engineering University
Priority date: 2024-01-16
Filing date: 2024-01-16
Publication date: 2024-02-20
Anticipated expiration: 2044-01-16
Also published as: CN117575371B

Abstract

The invention relates to the technical field of aerospace, and particularly discloses a multi-star imaging task planning method and system based on man-machine interaction, wherein the method comprises the following steps: constructing a multi-star imaging task planning model according to a multi-level objective function, a target task allocation constraint, a target visibility constraint and a satellite self-performance constraint of the multi-star imaging task planning; determining importance of a target task according to external comprehensive information of the satellite so as to generate an initial task set of a target task observation sequence; randomly generating a plurality of groups of multi-star imaging task distribution sequences according to the initial task set and determining an optimal pareto front surface; the individuals on the front surface of the optimal pareto are evaluated, the individuals are divided into different sub-populations according to the evaluation result, the population individuals of each sub-population are mapped into corresponding sub-target evaluation matrixes according to a mapping mechanism, and sub-target evaluation values are determined; receiving a demand instruction sent by a user, and generating a user preference planning scheme; and constructing a reflection mechanism and generating a task planning scheme.

Description

Multi-star imaging task planning method and system based on man-machine interaction

Technical Field

The invention relates to the technical field of aerospace, in particular to a multi-star imaging task planning method and system based on man-machine interaction.

Background

In recent years, the aerospace industry rapidly develops, the number of times of satellite transmission and the number of in-orbit satellites at home and abroad are increased year by year, and the capacity and range of the in-orbit satellites for executing imaging tasks are greatly expanded by multi-satellite cooperative cooperation. Therefore, the increasingly complex airspace environment and the user demands also provide higher requirements for the field of multi-satellite imaging task planning, the manual intervention is introduced for the complex and changed environment to quickly respond to the multi-user demands, and the generation of the multi-satellite imaging task planning scheme which is closer to the real demands is an unavoidable research problem in the field of aerospace technology application.

The multi-star system has the advantages of multi-azimuth continuous monitoring when the distribution area is wider and long-acting on the observation area can be realized due to the fact that the number of the distributed satellites is large, and plays an important role in the field of aerospace technology. The multi-satellite imaging task planning refers to the problem of large-scale combination optimization of the overall efficiency of the multi-satellite imaging task planning system to be optimal while the number of the objects to be observed is maximized by distributing one or a group of ordered objects to the imaging satellites according to multiple distribution requirements such as multi-user requirements, multi-satellite, multi-task and the like under a complex environment situation.

At present, in the aspect of application of a multi-star mission planning scheme, the following problems still exist:

1) The planning scheme output by the multi-star imaging task planning system needs to be finally applied to a real environment, however, the real application environment often has unknown complex factors due to complexity and variability, the traditional task planning method is often based on modeling and solving, and the difference between the model and the real environment is difficult to effectively solve. How to make the planning scheme more suitable for practical application demands by introducing human planning and decision-making capability in the traditional multi-star imaging task planning and searching technology is a great challenge for effective application of the multi-star imaging task planning scheme.

2) The multi-star imaging task planning problem is more complex in constraint under a complex environment, the problem space is more huge, the action effect uncertainty is stronger, the existing solving algorithm cannot adapt to various change modes, and solving performance and computing efficiency cannot be effectively balanced. In the process of optimizing the solution, the problem solving space is effectively reduced, and the construction of the multi-star imaging task planning scheme which can effectively adapt to various changes in a complex environment to obtain high quality is a new challenge of multi-star imaging task planning solution.

Therefore, how to solve the problems existing in the prior art is a urgent need for a person skilled in the art.

Disclosure of Invention

In order to achieve the purpose of the invention, the application provides a multi-star imaging task planning method based on man-machine interaction, which comprises the following steps:

step S1: constructing a multi-star imaging task planning model according to a multi-level objective function, a target task allocation constraint, a target visibility constraint and a satellite self-performance constraint of the multi-star imaging task planning;

step S2: determining importance of a target task according to external comprehensive information of the satellite so as to generate an initial task set of a target task observation sequence;

step S3: randomly generating a plurality of groups of multi-star imaging task distribution sequences according to the initial task set and determining an optimal pareto front surface;

step S4: the individuals on the front surface of the optimal pareto are evaluated, the individuals are divided into a task benefit sub-population, a task timeliness sub-population and a task completion degree sub-population according to the evaluation result, population individuals of each sub-population are mapped into corresponding sub-target evaluation matrixes according to a mapping mechanism, and sub-target evaluation values are determined;

step S5: receiving a demand instruction sent by a user, and generating a user preference planning scheme according to the demand instruction;

Step S6: and constructing a reflection mechanism, and carrying out reflection according to the user preference planning scheme based on the demand instruction to generate a final task planning scheme.

In some embodiments, the multi-stage objective function of the multi-star imaging mission plan includes a mission benefit objective function, a mission timeliness objective function, and a mission completion objective function; dividing the target task into important tasks and sub-important tasks according to the importance degree of the target task; selecting a target function in a self-adaptive mode according to the target task attribute, and selecting a task benefit target function from a task set with the secondary important task weight higher than that of the important task; otherwise, selecting a task timeliness objective function.

In some embodiments, the constraint includes: target task allocation constraints, target visibility constraints, and satellite own performance constraints.

In some embodiments, in step S2, the external integrated information of the satellite includes: environmental information, target task information and satellite state information; wherein,

the environment information comprises the influence of weather condition changes on the in-orbit running satellite, the influence of solar storm in the space environment and the influence of space debris impact;

The target task information comprises user demand change information, task execution importance change information and task demand resolution change information;

the satellite self state information comprises satellite load failure information, satellite power supply failure information and internal circuit abnormality information which appear in the process of executing tasks.

In some embodiments, the step S2 includes: the target task importance is generated by adopting a rolling type mixed priority method, the rolling type mixed priority method comprises a fixed priority mode and a dynamic priority mode, wherein the fixed priority mode is adopted to generate the target task importance at the initial planning moment, and if the target task importance is triggered by time and events, the dynamic priority mode is started according to the external comprehensive information of the satellite to update the target task importance correspondingly.

In some embodiments, the step S3 includes:

step S31: randomly generating a plurality of groups of multi-star imaging task allocation sequences serving as an initial parent population according to the initial task set;

step S32: solving the multi-star imaging task planning model by adopting an NSGA-II algorithm, and rapidly obtaining an initial pareto front surface of the initial parent population by a non-dominant sorting method;

Step S33: and optimizing the initial pareto front by a reference point decomposition method, and generating an optimal pareto front so as to obtain an initial allocation reference knowledge base.

In some specific embodiments, the step S4 includes:

step S41: respectively sequencing the fitness values of the individuals on the optimal pareto front surface by taking the self-fitness values of the task benefit objective function, the task timeliness objective function and the task completion objective function as evaluation standards;

step S42: sorting according to the fitness value descending order based on the task benefit objective function to generate a task benefit sub-population, sorting according to the fitness value descending order based on the task timeliness objective function to generate a task timeliness sub-population, and sorting according to the fitness value descending order based on the task completion objective function to generate a task completion sub-population;

step S43: a mapping mechanism is constructed, and population individuals of the task benefit sub-population, the task timeliness sub-population and the task completion degree sub-population are mapped into a task benefit evaluation matrix, a task timeliness evaluation matrix and a task completion degree evaluation matrix respectively;

step S44: and constructing a target task evaluation knowledge base to store the task benefit evaluation matrix, the task timeliness evaluation matrix and the task completion degree evaluation matrix.

In some embodiments, in step S5, the demand instruction includes one or more of a task benefit priority instruction, a task completion priority instruction, and a response speed priority instruction; and the demand instruction includes a single instruction, a dual instruction, or a multiple instruction.

In some embodiments, the sub-objective evaluation values are demapped according to a demapping mechanism to obtain population individuals, and a task planning scheme is generated according to the demand instructions.

In order to achieve the same purpose, the application also provides a multi-star imaging task planning system based on man-machine interaction, which comprises the following steps:

model construction module: the multi-satellite imaging task planning model is constructed according to a multi-level objective function, a target task allocation constraint, a target visibility constraint and a satellite self-performance constraint of the multi-satellite imaging task planning;

and a pretreatment module: the method comprises the steps of determining importance of a target task according to external comprehensive information of a satellite so as to generate an initial task set of a target task observation sequence;

a pre-planning module: the method comprises the steps of randomly generating a plurality of groups of multi-star imaging task distribution sequences according to an initial task set and determining an optimal pareto front surface;

The allocation reference module: the method comprises the steps of evaluating individuals on the front surface of the optimal pareto, dividing the individuals into a task benefit sub-population, a task timeliness sub-population and a task completion degree sub-population according to an evaluation result, mapping population individuals of each sub-population into corresponding sub-target evaluation matrixes according to a mapping mechanism, and determining sub-target evaluation values;

preference interaction module: the method comprises the steps of receiving a demand instruction sent by a user and generating a user preference planning scheme according to the demand instruction;

and the interaction output module is used for: and the method is used for constructing a reflection mechanism, and carrying out reflection according to the user preference planning scheme based on the demand instruction to generate a final task planning scheme.

The beneficial effects of the technical scheme are that:

1) According to the invention, human interactive information is introduced, and through manual intervention of the preprocessing module and the preference interaction module, preference input of single instruction, double instruction and multiple instruction is provided for the system, so that the gap between a planning model and reality is effectively reduced, the planning and searching technology can better serve the practical application environment, the corresponding model and algorithm are more accurate and more efficient, and meanwhile, the actual application requirements of people are more met, and a new research thought is provided for the traditional planning and searching technology.

2) According to the invention, by constructing the pre-planning module and the distribution reference module, an initial distribution reference knowledge base and a target task evaluation knowledge base are provided for the preference demands of the satellite management and control tasks under different scene demands, a mapping mechanism and a reflection mechanism are constructed, the sub-target evaluation matrix is adopted to replace the execution satellite and the target task sequence to participate in optimized search, the solving space of the problem is effectively reduced, and the multi-satellite multi-task can be ensured to rapidly plan a planning scheme more suitable for the current environment.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings that are needed in the description of the embodiments will be briefly introduced below, it being obvious that the drawings in the following description are only some embodiments of the present application, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

Fig. 1 is a schematic flow chart of a multi-star imaging task planning method based on man-machine interaction according to an embodiment of the present invention;

FIG. 2 is a schematic structural diagram of a multi-star imaging mission planning system according to an embodiment of the present invention;

FIG. 3 is a schematic diagram of a specific structure of a multi-star imaging task planning method and system based on man-machine interaction according to an embodiment of the present invention;

FIG. 4 is a constraint diagram related to a multi-star imaging mission planning model constructed based on a man-machine interaction type multi-star imaging mission planning method and system according to an embodiment of the present invention;

FIG. 5 is a schematic diagram of a single instruction, double instruction, and multiple instruction next generation population individuals based on a method and system for planning a multi-star imaging task according to one embodiment of the present invention;

FIG. 6 is a schematic diagram illustrating an example of a single demand instruction for a multi-star imaging task planning method and system based on man-machine interaction according to an embodiment of the present invention;

fig. 7 is a schematic diagram of an example in which a demand instruction of a multi-star imaging task planning method and system based on man-machine interaction is a single instruction according to an embodiment of the present invention.

Detailed Description

The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments.

Examples of the embodiments are illustrated in the accompanying drawings, wherein like or similar symbols indicate like or similar elements or elements having like or similar functions throughout. The embodiments described below by referring to the drawings are illustrative and intended to explain the present invention and should not be construed as limiting the invention.

Example 1

An embodiment of the present invention provides a method for planning a multi-star imaging task based on man-machine interaction, referring to fig. 1, including:

step S1: and constructing a multi-star imaging task planning model according to a multi-level objective function, a target task allocation constraint, a target visibility constraint and a satellite self-performance constraint of the multi-star imaging task planning.

In a specific embodiment of the present invention, the multi-stage objective function of the multi-star imaging task plan includes a task benefit objective function, a task timeliness objective function, and a task completion objective function; dividing the target task into important tasks and sub-important tasks according to the importance degree of the target task; selecting a target function in a self-adaptive mode according to the target task attribute, and selecting a task benefit target function from a task set with the secondary important task weight higher than that of the important task; otherwise, selecting a task timeliness objective function.

The multi-level objective function is as follows:

first sub-objective function: task benefit objective functionf ₁ ：

；

Wherein,Taskrepresenting a set of target tasks,Satrepresenting a set of satellites in view,Orbrepresenting the set of orbits performed by the satellite,Wa target importance matrix representing the target task,x _i,j,o the decision factor is represented by a number of factors,w _i representing target tasksiIs provided with a target importance level of (1),N _Task representing the number of tasks to be targeted,N _Sat representing the number of satellites in view,N _Orb representing the number of executable track turns.

Second sub-objective function: task timeliness objective functionf ₂ ：

；

In,representing target tasksiIs a response time of observation of (a);

third sub-objective function: task completion degree objective functionf ₃ ：

；

As shown in fig. 4, in some of these embodiments, the constraint includes: target task allocation constraints, target visibility constraints, and satellite own performance constraints.

Target task allocation constraints: each satellite may observe multiple target tasks in turn, each of which may be observed by multiple satellites.

；

SA _j Represent the firstjThe satellite performs the firstiThe probability of success of the observation of the individual target tasks;SA _j represent the firstjCumulative probability of the individual satellite performing the target;x _ij represent the firstjThe satellite performs the firstiDecision variables of the individual target tasks;

target visibility constraints: the satellite is visible to the target mission in the elevation intersection depression angle interval.

；

Wherein,representing the elevation-intersection depression angle of the imaging satellite when perpendicular to the target mission,Rrepresenting the linear distance of the target task from the centroid,hrepresenting the linear distance of the imaging satellite from the ground, +.>Representing the visible angle of the imaging satellite to the target mission.

Satellite own performance constraints:

；

wherein,indicating that the satellite is in orbitOResource consumption per unit time, +.>Representing satellitesjOn the trackOExecuting target tasks oniIs energy-consuming and is->、/>Respectively represent satellitesjOn the trackOExecuting target tasks oniIs (are) observed end and start times, /)>Representing the satellite's own energy.

Step S2: and determining the importance degree of the target task according to the external comprehensive information of the satellite so as to generate an initial task set of the target task observation sequence.

In a specific embodiment of the present invention, in step S2, the external integrated information of the satellite includes: environmental information, target task information and satellite state information; wherein,

In a specific embodiment of the present invention, the step S2 includes:

step S21, the environmental information mainly considers dynamic factors affecting the execution of multi-star imaging task planning, including: the on-orbit operation satellite is influenced by weather condition change and is influenced by solar storm and space debris impact in the space environment; the target task information mainly considers the change of the user demand, the change of the importance degree of task execution and the change of the resolution of the task demand; the satellite self state information mainly considers the interference of uncertain factors of satellite load failure, satellite power failure and internal circuit abnormality of the satellite in the task execution process;

and S22, generating the target task importance by adopting a rolling type mixed priority method, wherein the rolling type mixed priority method comprises a fixed priority mode and a dynamic priority mode, wherein the fixed priority mode is adopted to generate the target task importance at the initial planning moment, and then if the target task importance is triggered by time and an event, the dynamic priority mode is started to correspondingly update the target task importance according to the external comprehensive information of the satellite.

Preferably, step S22 specifically includes:

in step S221, the target task importance calculation formula is generated by adopting the rolling type mixed priority method as follows:

；

wherein,w _i representing target tasksiIs provided with a target importance level of (1),D _ij representing target tasksiFirst, thejThe quantized values of the individual indicators are used,Wa target importance matrix representing the target task,N _task representing the total number of tasks to be targeted,representing an initial set of tasks;

step S222, setting a single planning period of the multi-star imaging task planning system to 24 hours, and taking time and event triggering conditions into consideration, wherein the periodic rolling time of the rolling type mixed priority method is as followsTIn addition, at any time in the planning period, the system receives the change from the external comprehensive information, the new task planning requirement and the change of the planning task occur, the rolling type mixed priority work is triggered, the importance of each target task is dynamically changed,t+the starting time of the rolling type mixed priority at the moment 1 is as follows:

；

wherein,et _t for the time of event triggering, the triggering event mainly considers the emergency event which makes the satellite unable to execute the task according to the original plan, including but not limited to the satellite being in the processSatellite load failure, satellite power failure, abnormal internal circuit, cloud cover shielding, user demand change and task cancellation occur in the task execution process. T _t Representation oftStarting time of the time rolling type mixed priority method;

in step S223, by the rolling type mixed priority method, the satellite management and control personnel can update the external comprehensive information according to the actual requirement, so as to implement the intervention of the target task observation sequence.

Step S3: and randomly generating a plurality of groups of multi-star imaging task distribution sequences according to the initial task set and determining the optimal pareto front surface.

In a specific embodiment of the present invention, the step S3 includes:

step S33: optimizing the initial pareto front surface by a reference point decomposition method, generating an optimal pareto front surface to obtain an initial allocation reference knowledge base, wherein the unit knowledge of the initial allocation reference knowledge base is formed by the following steps ofSpecifically comprises executing satellite number sequence, target task sequence, task benefit, task response time and task completion degree 5, and initially distributing reference knowledge base Pool _allocation The expression of (2) is as follows:

specifically, the initial task contains 6 target tasks, specifically、/>、、/>、/>、/>Finally, unit knowledge of the initial allocation reference knowledge base is obtained as follows:

。

step S4: and evaluating individuals on the optimal pareto front surface, dividing the individuals into a task benefit sub-population, a task timeliness sub-population and a task completion degree sub-population according to an evaluation result, mapping population individuals of each sub-population into corresponding sub-target evaluation matrixes according to a mapping mechanism, and determining sub-target evaluation values.

Step S41: respectively sequencing the fitness values of three sub-targets of the individual on the optimal pareto front surface by taking the respective fitness values of the task benefit objective function, the task timeliness objective function and the task completion degree objective function as evaluation standards;

taking the unit knowledge shown in the initial allocation reference knowledge base as an example, the individual units areSequencing the fitness values of three sub-targets of individuals on the optimal pareto front surface, and sequencing the fitness values of the task benefit objective function, wherein the individuals are +.>The method comprises the steps of carrying out a first treatment on the surface of the Ordering by task timeliness objective function fitness value, wherein the individual is +.>The method comprises the steps of carrying out a first treatment on the surface of the Ordering by task completion degree objective function fitness value, individual is +. >；

Step S42: generating task benefit sub-population based on task benefit objective function according to descending order of fitness valueP ₁ Generating a task timeliness sub-population based on the task timeliness objective function in descending order of fitness valueP ₂ Generating a task completion degree sub-population based on a task completion degree objective function in descending order of fitness valuesP ₃ ；

Step S43: constructing a mapping mechanism to group task benefitsP ₁ Time-efficient sub-population of tasksP ₂ Task completion degree sub-populationP ₃ Are respectively mapped into task benefit evaluation matrixTask timeliness evaluation matrixTask completion evaluation matrix->；

The mapping mechanism maps the execution satellite of the discrete evaluation space and the target task sequence to the continuous scheme evaluation space through the corresponding value of each sub-target, and the mapping mechanism is as follows:

according to the task benefit sub-populationP ₁ Constructing a task benefit evaluation matrix by taking a current individual (an allocation sequence for executing a satellite and a target task) as a mapping relationThe method comprises the steps of carrying out a first treatment on the surface of the Time-dependent sub-populations of the taskP ₂ Constructing a task timeliness evaluation matrix>The method comprises the steps of carrying out a first treatment on the surface of the According to the task completion degreePopulation groupP ₃ Constructing a task completion degree evaluation matrix。

Step S44: building a target task assessment knowledge base to store the task benefit assessment matrix Task timeliness assessment matrix>Task completion evaluation matrix->The method comprises the steps of carrying out a first treatment on the surface of the In the optimized search, the evaluation matrix of each sub-target is required to be directly called aiming at the user preference, and a reference is provided for final planning allocation.

The target task evaluation knowledge base generated by the embodiment specifically comprises the following steps:

task benefit evaluation matrix:

；

task timeliness assessment matrix:

；

task completion degree evaluation matrix:

。

step S5: and receiving a demand instruction sent by a user, and generating a user preference planning scheme according to the demand instruction.

In one embodiment of the present invention, the demand instruction includes one or more of a task benefit priority instruction, a task completion priority instruction, and a response speed priority instruction; and the demand instruction includes a single instruction, a dual instruction, or a multiple instruction.

Step S51: providing experience knowledge for a system sending demand instruction according to actual application demands, wherein the system sending instruction comprises: task benefit priority instruction, task completion priority instruction, response speed priority instruction; the system sending instruction can send a single instruction, a double instruction and multiple instructions according to actual requirements;

FIG. 5 is a schematic diagram of a next generation population of single instruction, double instruction, and multiple instruction;

Step S52: if the demand instruction is a single instruction, carrying out preference selection on a new generation population according to the system sending instruction, wherein the new generation population is constructed in the following way:

；

wherein,P _new represents a new generation of population, and the method comprises the steps of,Prepresenting the initial parent population of the parent,P _i representing sub-populations, in particular task benefit sub-populationsP ₁ Time-efficient sub-population of tasksP ₂ Task completion degree sub-populationP ₃ ；

Specifically, taking a system sending instruction as a task benefit priority instruction as an example, the new generation population isThe method comprises the steps of carrying out a first treatment on the surface of the Taking the satellite management and control personnel as an example to send response speed priority instructions, the new generation population isThe method comprises the steps of carrying out a first treatment on the surface of the Taking the satellite management and control personnel as an example to send task completion priority instructions, the new generation population is。

Step S53: if the satellite management and control personnel send a double instruction, carrying out preference selection on a new generation population according to the double instruction, wherein the new generation population is constructed in the following way:

；

wherein, the single instruction of sending is unrepeatable, and the combination mode of dual instruction includes: task benefit priority instruction and task completion priority instruction, task benefit priority instruction and response speed priority instruction, task completion priority instruction and response speed priority instruction.

Specifically, taking the satellite management and control personnel to send the task benefit priority instruction and the task completion priority instruction as examples, the new generation population is The method comprises the steps of carrying out a first treatment on the surface of the Taking the satellite management and control personnel as an example to send task benefit priority instructions and response speed priority instructions, the new generation population is +.>The method comprises the steps of carrying out a first treatment on the surface of the Taking the satellite management and control personnel as an example of sending task completion priority instructions and response speed priority, the new generation population is +.>。

Step S54: if the satellite management and control personnel send multiple instructions, the multi-star imaging task planning system is required to output a multi-star imaging task planning scheme which simultaneously meets the priority of task benefit, response speed and task completion, and the new generation population construction mode is as follows:

；

it is now optimized as a non-biased multi-objective problem.

Preferably, the step S54 specifically includes:

the three demand targets of the priority of task benefit, the priority of response speed and the priority of task completion are expanded to a uniform magnitude through expansion coefficients, and the specific expression is as follows:

；

in order to be a benefit of the task,βfor response time +.>In order to achieve the degree of completion of the task,c _i is an expansion coefficient against the constraint.

Step S55: and carrying out evolution operation according to the new generation population construction to generate a user preference planning scheme so as to realize dynamic adjustment of the user preference task planning scheme in the optimized search process, wherein the evolution operation comprises crossover, mutation and selection operation.

In a specific embodiment of the invention, the sub-objective evaluation values are reversely mapped according to a reflection mechanism to obtain population individuals, and a task planning scheme is generated according to the demand instructions.

Step S61: if the satellite management and control personnel only send a single instruction, the reflection mechanism carries out one-to-one matching on the current sub-target evaluation value obtained by the evolution operation and the historical sub-target evaluation value of the corresponding evaluation matrix in the target task evaluation knowledge base, matches the historical sub-target evaluation value with the smallest difference value with the current sub-target evaluation value in the evaluation matrix, and reflects the historical sub-target evaluation value to a corresponding sequence relation, and the reflection mechanism is as follows:

wherein,、/>、/>respectively represent the current satellitesjExecuting target tasksiTask benefit value, task response time, task completion; />The method of reflection is represented by a method of reflection,、/>、/>respectively represent satellites in a target task evaluation knowledge basejExecuting target tasksiCorresponding historical task benefit value, historical task response time and historical task completion degree;

step S62: if the satellite management and control personnel send a double instruction, the reflection mechanism converts two current evaluation matrixes obtained by evolution operation into a current average matrix, converts a corresponding evaluation matrix in a target task evaluation knowledge base into a historical average matrix, performs one-to-one matching on sub-target evaluation values of the current average matrix and the historical average matrix, matches out a historical sub-target evaluation value with the smallest difference value of the sub-target evaluation values, and reflects and shoots out a corresponding sequence relation, and the reflection mechanism is as follows:

Step S63: aiming at the single instruction and double instruction conditions, sub-target evaluation values are reversely mapped according to the reflection mechanism to obtain population individuals, and the system outputs a plurality of multi-star imaging task planning schemes for satellite control personnel to select; the multiple instruction case goes to step S64.

Taking a task benefit priority instruction as an example, a single instruction is sent to satellite management and control personnel, a new generation population is evolved, and a satellite and a target sequence are obtained by reflection and reflection for explanation. FIG. 6 shows an example of a task benefit priority instruction sent by a satellite manager, the new generation population being obtained by an evolutionary operationThe task benefit target evaluation matrix is:invoking task benefit assessment matrix->As a reflection reference.

The current sub-target evaluation value 191 is matched with each historical sub-target evaluation value in the first column of the task benefit evaluation matrix one by one, the historical sub-target evaluation value 186 with the smallest difference value with the current sub-target evaluation value is matched in the evaluation matrix, and reflected to obtain the satellite and target task sequence relation asAfter that, the rows and columns of the historical sub-target evaluation values 186 are all invalid values, the invalid values are deleted from the evaluation matrix, and the next sub-target evaluation value 405 is matched; sub-target evaluation value 405 reflects the satellite and target task sequence relationship of +. >Similarly, sub-target evaluation value 156 is reflected to obtain satellite and target task sequence relationship of +.>Sub-target evaluation value 323 reflects the satellite and target task sequence relationship of +.>Sub-target evaluation value 334 is reflected to obtain satellite and target task sequence relation as followsSub-target evaluation value 417, reflected in satellite and target mission sequence relation of +.>。

Taking the case of sending a task benefit priority instruction and a response speed priority instruction as an example, the output corresponding sequence relationship of the reflection mechanism is described, as shown in fig. 7, which is an example of sending a task benefit priority instruction by a satellite management and control personnel, and the history average matrix is as follows:

；

the task benefit target evaluation matrix obtained by the evolution operation of the new generation population is as follows:the response time target evaluation matrix is:the current average matrix is: />The method comprises the steps of carrying out a first treatment on the surface of the The current average matrix and the historical average matrix are matched one by one, and the satellite and target task sequences obtained through reflection are as follows:、/>、/>、/>、/>、/>。

step S64: aiming at the condition of multiple instructions, the multi-star imaging task planning system selects a planning scheme with the maximum fitness value as an optimal scheme to output.

The invention provides the preference input of single instruction, double instruction and multiple instruction for the system, effectively reduces the gap between the planning model and reality, thereby enabling the planning and searching technology to better serve the practical application environment, enabling the corresponding model and algorithm to be more accurate and more efficient, simultaneously being more in accordance with the practical application demands of people, and providing a new research thought for the traditional planning and searching technology.

The invention also provides an initial allocation reference knowledge base and a target task evaluation knowledge base for the preference requirements of the satellite management and control tasks under different scene requirements, builds a mapping mechanism and a reflecting mechanism, adopts a sub-target evaluation matrix to replace an execution satellite and a target task sequence to participate in optimized search, effectively reduces the solving space of the problem, and ensures that multi-satellite multi-task can be rapidly planned to be more suitable for the planning scheme of the current environment.

Example two

One embodiment of the present invention provides a man-machine interaction-based multi-star imaging mission planning system, as shown in fig. 2 and 3, including:

model building module 10: the multi-star imaging task planning model is used for constructing a multi-star imaging task planning model according to a multi-level objective function, a target task allocation constraint, a target visibility constraint and a satellite self-performance constraint of the multi-star imaging task planning.

In one embodiment of the invention, a multi-stage objective function of multi-star imaging task planning is constructed, wherein the multi-stage objective function comprises a task benefit objective function, a task timeliness objective function and a task completion objective function; dividing the target task into important tasks and sub-important tasks according to the importance degree of the target task; selecting a target function in a self-adaptive mode according to the target task attribute, and selecting a task benefit target function from a task set with the secondary important task weight higher than that of the important task; otherwise, selecting a task timeliness objective function.

The multi-level objective function is as follows:

first sub-objective function: task benefit objective functionf ₁ ：

；

Second sub-objective function: task timeliness objective functionf ₂ ：

；

In,representing target tasksiIs a response time of observation of (a);

；

Pretreatment module 20: and the initial task set is used for determining the importance degree of the target task according to the external comprehensive information of the satellite so as to generate the target task observation sequence.

In one embodiment of the present invention, the external integrated information of the satellite includes: environmental information, target task information and satellite state information; wherein,

In one embodiment of the present invention, the preprocessing module 20 is further configured to:

an information processing unit: the environmental information is determined according to dynamic factors of multi-star imaging task planning execution, and the method comprises the following steps: the on-orbit operation satellite is influenced by weather condition change and is influenced by solar storm and space debris impact in the space environment; the target task information is determined according to the user demand change, the task execution importance change and the task demand resolution change information; the satellite self state information mainly considers the interference of uncertain factors of satellite load failure, satellite power failure and internal circuit abnormality of the satellite in the task execution process;

importance degree determination unit: the target task importance is generated by adopting a rolling type mixed priority method, the rolling type mixed priority method comprises a fixed priority mode and a dynamic priority mode, wherein the fixed priority mode is adopted to generate the target task importance at the initial planning moment, and if the target task importance is triggered by time and events, the dynamic priority mode is started according to the external comprehensive information of the satellite to update the target task importance correspondingly.

Preferably, the importance determining unit is specifically configured to:

the calculation formula for generating the importance degree of the target task by adopting the rolling type mixed priority method is as follows:

；

the single planning period of the multi-star imaging task planning system is set to be 24 hours, and the time and event triggering conditions are considered at the same time, and the periodic rolling time of the rolling type mixed priority method is as followsTIn addition, at any time in the planning period, the system receives the change from the external comprehensive information, the new task planning requirement and the change of the planning task occur, the rolling type mixed priority work is triggered, the importance of each target task is dynamically changed,t+the starting time of the rolling type mixed priority at the moment 1 is as follows:

；

wherein,et _t for the event triggering time, the triggering event mainly considers the emergency event that the satellite cannot execute the task according to the original plan, including but not limited to satellite load failure, satellite power failure, internal circuit abnormality, cloud cover, user demand change and task cancellation. T _t Representation oftStarting time of the time rolling type mixed priority method;

by the rolling type mixed priority method, satellite management and control personnel can update external comprehensive information according to actual requirements so as to realize intervention on the observation sequence of the target task.

Pre-planning module 30: and the method is used for randomly generating a plurality of groups of multi-star imaging task distribution sequences according to the initial task set and determining the optimal pareto front.

In one embodiment of the present invention, the pre-planning module 30 includes:

an initial parent population generation unit: randomly generating a plurality of groups of multi-star imaging task allocation sequences serving as an initial parent population according to the initial task set;

initially pareto front generating unit: solving the multi-star imaging task planning model by adopting an NSGA-II algorithm, and rapidly obtaining an initial pareto front surface of the initial parent population by a non-dominant sorting method;

initial allocation reference knowledge base generation unit: optimizing the initial pareto front surface by a reference point decomposition method, generating an optimal pareto front surface to obtain an initial allocation reference knowledge base, wherein the unit knowledge of the initial allocation reference knowledge base is formed by the following steps ofSpecifically comprises executing satellite number sequence, target task sequence, task benefit, task response time and task completion degree 5, and initially distributing reference knowledge base Pool _allocation The expression of (2) is as follows:

；/>

。

allocation reference module 40: the method comprises the steps of evaluating individuals on the optimal pareto front surface, dividing the individuals into a task benefit sub-population, a task timeliness sub-population and a task completion degree sub-population according to an evaluation result, mapping population individuals of each sub-population into corresponding sub-target evaluation matrixes according to a mapping mechanism, and determining sub-target evaluation values.

And the fitness value sequencing unit is used for: respectively sequencing the fitness values of three sub-targets of the individual on the optimal pareto front surface by taking the respective fitness values of the task benefit objective function, the task timeliness objective function and the task completion degree objective function as evaluation standards;

A descending order sorting unit: generating task benefit sub-population based on task benefit objective function according to descending order of fitness valueP ₁ Generating a task timeliness sub-population based on the task timeliness objective function in descending order of fitness valueP ₂ Task completion degree-based goalThe functions are ordered according to the descending order of fitness value to generate a task completion degree sub-populationP ₃ ；

Mapping mechanism construction unit: constructing a mapping mechanism to group task benefitsP ₁ Time-efficient sub-population of tasksP ₂ Task completion degree sub-populationP ₃ Are respectively mapped into task benefit evaluation matrixTask timeliness assessment matrix>Task completion evaluation matrix->；

according to the task benefit sub-populationP ₁ Constructing a task benefit evaluation matrix by taking a current individual (an allocation sequence for executing a satellite and a target task) as a mapping relationThe method comprises the steps of carrying out a first treatment on the surface of the Time-dependent sub-populations of the taskP ₂ Constructing a task timeliness evaluation matrix>The method comprises the steps of carrying out a first treatment on the surface of the Sub-population according to the task completion degreeP ₃ Constructing a task completion degree evaluation matrix>。

The target task evaluation knowledge base construction unit: building a target task assessment knowledge base to store the task efficiency Benefit evaluation matrixTask timeliness assessment matrix>Task completion evaluation matrix->The method comprises the steps of carrying out a first treatment on the surface of the In the optimized search, the evaluation matrix of each sub-target is required to be directly called aiming at the user preference, and a reference is provided for final planning allocation.

task benefit evaluation matrix:

；

task timeliness assessment matrix:

；

task completion degree evaluation matrix:

。

preference interaction module 50: and the user preference planning scheme is used for receiving the demand instruction sent by the user and generating the user preference planning scheme according to the demand instruction.

An instruction transmitting unit: providing experience knowledge for a system sending demand instruction according to actual application demands, wherein the system sending instruction comprises: task benefit priority instruction, task completion priority instruction, response speed priority instruction; the system sending instruction can send a single instruction, a double instruction and multiple instructions according to actual requirements;

a single instruction population building unit: if the demand instruction is a single instruction, carrying out preference selection on a new generation population according to the system sending instruction, wherein the new generation population is constructed in the following way:

；

Specifically, taking a system sending instruction as a task benefit priority instruction as an example, the new generation population isThe method comprises the steps of carrying out a first treatment on the surface of the Taking the system sending instruction as a response speed priority instruction as an example, the new generation population isThe method comprises the steps of carrying out a first treatment on the surface of the Taking the system sending instruction as the task completion priority instruction as an example, the new generation population is。

A dual instruction population construction unit: if the demand instruction is a double instruction, selecting a new generation population in a preference mode according to the double instruction, wherein the new generation population is constructed in the following mode:

；

Specifically, taking a system sending instruction as a task benefit priority instruction and a task completion priority instruction as examples, the new generation population isThe method comprises the steps of carrying out a first treatment on the surface of the Taking a system sending instruction as a task benefit priority instruction and a response speed priority instruction as examples, the new generation population is +.>The method comprises the steps of carrying out a first treatment on the surface of the Taking a system sending instruction as a task completion degree priority instruction and a response speed priority as examples, the new generation population is +.>。

Multiple instruction population building unit: if the demand instruction is a multiple instruction, the multi-star imaging task planning system is required to output a multi-star imaging task planning scheme which simultaneously meets the priority of task benefit, response speed and task completion degree, and the new generation population construction mode is as follows:

；

it is now optimized as a non-biased multi-objective problem.

Preferably, the multiple instruction population construction unit specifically includes:

；

Preference planning scheme generation unit: and carrying out evolution operation according to the new generation population construction to generate a user preference planning scheme so as to realize dynamic adjustment of the user preference task planning scheme in the optimized search process, wherein the evolution operation comprises crossover, mutation and selection operation.

Interaction output module 60: and the method is used for constructing a reflection mechanism, and carrying out reflection according to the user preference planning scheme based on the demand instruction to generate a final task planning scheme.

Single instruction operation unit: if the demand instruction is a single instruction, the reflection mechanism carries out one-to-one matching on the current sub-target evaluation value obtained by the evolution operation and the historical sub-target evaluation value of the corresponding evaluation matrix in the target task evaluation knowledge base, matches the historical sub-target evaluation value with the smallest difference value with the current sub-target evaluation value in the evaluation matrix, and reflects the historical sub-target evaluation value to a corresponding sequence relation, and the reflection mechanism is as follows:

A dual instruction operation unit: if the demand instruction is a dual instruction, the de-mapping mechanism converts two current evaluation matrixes obtained by evolution operation into a current average matrix, converts a corresponding evaluation matrix in a target task evaluation knowledge base into a historical average matrix, performs one-to-one matching on sub-target evaluation values of the current average matrix and the historical average matrix, matches a historical sub-target evaluation value with a minimum difference value of the sub-target evaluation values, and reflects and maps out a corresponding sequence relation, and the de-mapping mechanism is as follows:

and a reflection unit: aiming at the single instruction and double instruction conditions, sub-target evaluation values are reversely mapped according to the reflection mechanism to obtain population individuals, and the system outputs a plurality of multi-star imaging task planning schemes for satellite control personnel to select; multiple instruction cases execute multiple instruction operation units.

Taking a task benefit priority instruction as an example, a demand instruction is a single instruction, a new generation population is evolved, and a satellite and a target sequence are obtained by reflection and are illustrated. FIG. 6 shows an example where the demand instruction is a task benefit priority instruction, and the task benefit target evaluation matrix obtained by the evolution operation of the new generation population is: Invoking task benefit assessment matrix->As a reflection reference.

The current sub-target evaluation value 191 is matched with each historical sub-target evaluation value in the first column of the task benefit evaluation matrix one by one, the historical sub-target evaluation value 186 with the smallest difference value with the current sub-target evaluation value is matched in the evaluation matrix, and reflected to obtain the satellite and target task sequence relation asAfter that, the rows and columns of the historical sub-target evaluation values 186 are all invalid values, the invalid values are deleted from the evaluation matrix, and the next sub-target evaluation value 405 is matched; sub-target evaluation value 405 reflects the satellite and target task sequence relationship of +.>Similarly, sub-target evaluation value 156 is reflected to obtain satellite and target task sequence relationship of +.>Sub-target evaluation value 323 reflects the satellite and target task sequence relationship of +.>Sub-target evaluation value 334 is reflected to obtain satellite and target task sequence relation as followsSub-target evaluation value 417, reflected in satellite and target mission sequence relation of +.>。

Taking the case of sending a task benefit priority instruction and a response speed priority instruction as an example, the output corresponding sequence relationship of the reflection mechanism is described, as shown in fig. 7, which shows an example that a demand instruction is a task benefit priority instruction, and the history average matrix is as follows:

；

The task benefit target evaluation matrix obtained by the evolution operation of the new generation population is as follows:the response time target evaluation matrix is:the current average matrix is:the method comprises the steps of carrying out a first treatment on the surface of the The current average matrix and the historical average matrix are matched one by one, and the satellite and target task sequences obtained through reflection are as follows: />、/>、/>、、/>、/>。

Multiple instruction operation unit: aiming at the condition of multiple instructions, the multi-star imaging task planning system selects a planning scheme with the maximum fitness value as an optimal scheme to output.

The foregoing is merely illustrative of the present invention, and the present invention is not limited thereto, and any person skilled in the art will readily recognize that variations or substitutions are within the scope of the present invention. Therefore, the protection scope of the invention is subject to the protection scope of the claims.

In this specification, each embodiment is described in a progressive manner, and each embodiment is mainly described by differences from other embodiments, and identical and similar parts between the embodiments are all enough to be referred to each other. Embodiments of the present invention are described with reference to flowchart illustrations and/or block diagrams of methods, terminal devices (systems), and computer program products according to embodiments of the invention. It will be understood that each flow and/or block of the flowchart illustrations and/or block diagrams, and combinations of flows and/or blocks in the flowchart illustrations and/or block diagrams, can be implemented by computer program instructions. These computer program instructions may be provided to a processor of a general purpose computer, special purpose computer, embedded processor, or other programmable data processing terminal device to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing terminal device, create means for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks. These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing apparatus to function in a particular manner, such that the instructions stored in the computer-readable memory produce an article of manufacture including instruction means which implement the function specified in the flowchart flow or flows and/or block diagram block or blocks. These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide steps for implementing the functions specified in the flowchart flow or flows and/or block diagram block or blocks. While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. It is therefore intended that the following claims be interpreted as including the preferred embodiment and all such alterations and modifications as fall within the scope of the embodiments of the invention. Finally, it is further noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or terminal that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or terminal. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article or terminal device comprising the element.

The foregoing has outlined rather broadly the more detailed description of the method and apparatus of the present invention in order that the detailed description of the principles and embodiments of the invention may be implemented in conjunction with the present invention; meanwhile, as those skilled in the art will have variations in the specific embodiments and application scope in accordance with the ideas of the present invention, the present description should not be construed as limiting the present invention in view of the above.

In the description of the present specification, reference to the terms "one embodiment," "some embodiments," "examples," "particular examples," "one particular embodiment," or "some examples," etc., means that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, schematic representations of terms do not necessarily refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.

Finally, it should be noted that: the above embodiments are only for illustrating the technical solution of the present application, and are not limiting thereof; although the present application has been described in detail with reference to the foregoing embodiments, one of ordinary skill in the art will appreciate that: the technical scheme described in the foregoing embodiments can be modified or some technical features thereof can be replaced by equivalents; such modifications and substitutions do not drive the essence of the corresponding technical solutions to depart from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A man-machine interaction-based multi-star imaging task planning method is characterized by comprising the following steps of:

step S1: constructing a multi-star imaging task planning model according to a multi-stage objective function and constraint conditions of the multi-star imaging task planning;

2. The man-machine interaction-based multi-star imaging task planning method according to claim 1, wherein the multi-stage objective functions of the multi-star imaging task planning comprise a task benefit objective function, a task timeliness objective function and a task completion objective function; the target tasks are divided into important tasks and sub-important tasks according to the importance degree of the target tasks.

3. The man-machine interaction-based multi-star imaging mission planning method of claim 1, wherein the constraints include: target task allocation constraints, target visibility constraints, and satellite own performance constraints.

4. The man-machine interaction-based multi-star imaging mission planning method as claimed in claim 1, wherein in step S2, the external integrated information of the satellite includes: environmental information, target task information and satellite state information; wherein,

5. The man-machine interaction-based multi-star imaging mission planning method of claim 1, wherein the step S2 comprises: the target task importance is generated by adopting a rolling type mixed priority method, the rolling type mixed priority method comprises a fixed priority mode and a dynamic priority mode, wherein the fixed priority mode is adopted to generate the target task importance at the initial planning moment, and if the target task importance is triggered by time and events, the dynamic priority mode is started according to the external comprehensive information of the satellite to update the target task importance correspondingly.

6. The man-machine interaction-based multi-star imaging mission planning method as claimed in claim 1, wherein the step S3 comprises:

7. The man-machine interaction-based multi-star imaging mission planning method of claim 2, wherein the step S4 comprises:

8. The man-machine interaction-based multi-star imaging task planning method according to claim 1, wherein in step S5, the demand instruction includes one or more of a task benefit priority instruction, a task completion priority instruction, and a response speed priority instruction; and the demand instruction includes a single instruction, a dual instruction, or a multiple instruction.

9. The man-machine interaction-based multi-star imaging mission planning method of claim 1, wherein in step S6, the sub-objective evaluation values are demapped according to a demapping mechanism to obtain population individuals, and a mission planning scheme is generated according to the demand instructions.

10. A man-machine interaction-based multi-star imaging mission planning system, comprising: