Knowledge Modeling and Transformation Method for Space Station Flight Control Task Domain Oriented to HTN Planning

Abstract

The present invention discloses a method for knowledge modeling and conversion in the field of space station flight control tasks oriented to HTN planning, constructs a knowledge concept model of a knowledge graph for space station flight control tasks oriented to HTN planning, and performs conversion between it and the knowledge model in the HTN planning field to generate the Domain file and Problem file required for HTN planning; this method can reduce the difficulty of constructing the planning domain model, improve the efficiency and effectiveness of model construction, provide effective domain knowledge model support for the development and execution of intelligent planning, and provide necessary Domain file and Problem file inputs for the HTN planner. Since the construction of the domain knowledge graph combines the planning experience of flight control domain experts and the actual flight control task planning process, it can be continuously updated and managed, and can guarantee the quality, completeness and effectiveness of domain knowledge.

Description

Technical Field

[0001] The present invention belongs to the field of space station mission planning, and more specifically, relates to a method for modeling and transforming domain knowledge of a space station flight control mission for HTN planning. Background Technique

[0002] After the space station enters the long-term operation stage from the on-orbit construction stage, mission planning has the complexity characteristics of complex mission structure, complex constraint conditions, highly dynamic operating environment, human-machine cooperation, etc. There is an urgent need to study intelligent planning methods to improve the operation efficiency of the space station and ensure the daily safe operation of the space station. The core of intelligent planning lies in the representation of domain knowledge, the construction of a planning domain model, and the writing of a planning domain file. Domain knowledge is an essential input for intelligent task planning algorithms and an important basis for task decomposition, method setting, and constraint handling by the planner. The ability, completeness, and effectiveness of domain knowledge representation will largely affect the efficiency of planning solution. As a widely used intelligent task planning method, HTN planning Figure 1 shows its conceptual model. The process of the planner's planning solution is to model the representation of domain knowledge as a planning domain knowledge Domain file and a planning domain problem Problem file as inputs, and output an action plan. Among them, the Domain file is a "prescription" and tool for planning problems within a specific domain, usually containing specific decomposition rules and execution rules, represented as a set of decomposition methods and a set of operators; the Problem file contains the system state and the task network, describing the real-time state information of the current environment and the task goals to be achieved; the action plan describes the action sequence that can be directly executed to complete the task goals, usually consisting of a series of instantiated operators with logical relationships. The representation and modeling of domain knowledge are the keys to supporting the practical application of intelligent planning technology. In current research on intelligent planning problems, most research is based on complete domain knowledge, and intelligent planning experts manually construct a planning domain knowledge model through planning domain modeling languages such as PDDL, thereby constructing a Domain file and a Problem file.

[0003] However, due to the complexity and uncertainty of real-world problems, etc., planning experts usually lack complete domain knowledge in related fields. Even if they have the ability to solve planning problems, they cannot construct a complete and high-quality planning domain knowledge model for complex practical problems. Although domain experts have in-depth understanding of domain knowledge, they usually lack professional knowledge and technology related to intelligent planning and it is difficult to represent domain knowledge as a planning domain knowledge model required for intelligent planning.

[0004] In the current research on spacecraft mission planning problems, most studies default to having complete domain knowledge and often directly construct domain knowledge models using planning domain modeling languages such as PDDL. This method is commonly used in satellite mission planning and deep space probe mission planning. For space station mission planning, foreign research mainly focuses on the description of software systems and lacks a detailed introduction to the modeling process; domestic research mainly proposes corresponding modeling methods, heuristic planning strategies, optimization algorithms, etc. from the perspective of operational mission planning. The International Space Station adopts an operation and management method of hierarchical planning, distributed collaboration, and centralized management. Countries and organizations such as the United States have developed a variety of mission planning systems, such as the core planning systems: the ISP integrated planning system of the Johnson Space Center and the PPS payload mission planning system of the Marshall Space Flight Center, etc., which can handle complex constraint conditions, orchestrate multiple resources, and meet on-orbit activities and ground requirements. However, most of the research focuses on the functional description of software systems and lacks an introduction to specific planning models and modeling methods.

[0005] At the same time, existing research does not fully consider the hierarchical decomposition relationship and complex constraint conditions in the space station flight control mission planning process, and has strong limitations. Due to the complex characteristics of space station flight control mission planning, such as multi-department collaboration, multi-task requirements, and multi-constraint coupling, the types, quantities, relationships, and uncertainties of the domain knowledge in the space station flight control mission planning are numerous. It is very difficult for ground control personnel to obtain complete domain knowledge and manually construct a planning domain model using a planning domain modeling language such as PDDL; the existing research on ontology-based space station knowledge modeling methods constructs problem models for constraint satisfaction theory and timeline theory, simplifies the concepts and constraints in the on-orbit mission planning domain, does not consider the hierarchical decomposition relationship between flight control tasks, lacks the description and representation of domain knowledge such as the TT&C network system and domain rules, and lacks the construction of a space station flight control mission domain knowledge model for HTN planning from the perspective of intelligent planning and in combination with the model requirements of intelligent planning.

[0006] The domain knowledge model conversion method is a key element in the combined application of knowledge modeling techniques such as ontologies and intelligent planning techniques, aiming to automatically generate the planning domain model required for intelligent planning and solve the problems that it is extremely difficult and not easy to modify the manually constructed planning domain model. Existing knowledge model conversion methods are mostly targeted at specific application domains, such as the robotics domain and the emergency domain, etc. The domain knowledge and planning knowledge of the application domain are combined to construct a knowledge concept model, and a conversion method related to the specific domain is designed according to the mapping relationship between the knowledge concept model and the planning domain model. Currently, for the space station flight control task planning problem, there is a lack of the mapping relationship between the flight control task domain knowledge concept model and the HTN planning domain knowledge model, and there is a lack of corresponding domain knowledge model conversion algorithms to automatically generate the domain knowledge file (Domain file) and domain problem file (Problem file) required for the space station flight control task planning oriented to HTN. Summary of the Invention

[0007] Aiming at the above defects or improvement requirements of the prior art, the present invention provides a method for modeling and converting the domain knowledge of the space station flight control task oriented to HTN planning, thereby solving the technical problem that it is difficult to directly construct the domain knowledge model of the space station flight control task planning using traditional planning domain modeling languages such as PDDL, and can provide a friendly and convenient modeling method for ground control personnel, represent, model, share and effectively manage the flight control task domain knowledge, realize the automatic reading and automatic conversion of domain knowledge, and provide domain knowledge model support for the subsequent research of intelligent planning methods.

[0008] To achieve the above object, according to the first aspect of the present invention, there is provided a method for modeling and converting the domain knowledge of the space station flight control task oriented to HTN planning, including:

[0009] S1, according to the core concepts, object attributes and data attributes of the space station flight control task planning domain knowledge, create the model ontology of the domain knowledge graph, perform knowledge processing on the space station flight control task planning data to obtain the basic elements of the domain knowledge graph and fill them into the model ontology to obtain the target model; wherein, the model includes a basic domain knowledge model and a planning domain knowledge model;

[0010] S2, respectively establish the mapping relationships between the planning domain knowledge model and the Domain file of the HTN planning domain model, and between the basic domain knowledge model and the Problem file of the HTN planning domain model;

[0011] S3. Read the flight control action instances and their attributes, and the relationships with other instances in the planned domain knowledge model. According to the mapping relationship, convert the flight control action instances into operators; read the flight control fragment instances and their attributes, and the relationships with other instances in the planned domain knowledge model. According to the mapping relationship, convert the flight control fragment instances into methods; add the obtained operator set and method set to the Domain file to obtain the target Domain file.

[0012] S4. Read each instance and its attributes in the basic domain knowledge model respectively, and convert them into states according to the mapping relationship; read the monthly event instances and their attributes in the basic domain knowledge model, and convert them into tasks according to the mapping relationship; add the obtained state set and task set to the Problem file to obtain the target Problem file.

[0013] According to the second aspect of the present invention, there is provided a space station flight control task planning method based on HTN planning. Input the target Domain file and the target Problem file obtained by using the method described in the first aspect into the HTN planner to obtain a space station flight control task planning scheme.

[0014] According to the third aspect of the present invention, there is provided a space station flight control task planning domain knowledge modeling and conversion system, which is characterized by including: a computer-readable storage medium and a processor;

[0015] The computer-readable storage medium is used to store executable instructions;

[0016] The processor is used to read the executable instructions stored in the computer-readable storage medium and execute the method described in the first aspect.

[0017] According to the fourth aspect of the present invention, there is provided a space station flight control task planning system based on HTN planning, including: a computer-readable storage medium and a processor;

[0018] The computer-readable storage medium is used to store executable instructions;

[0019] The processor is used to read the executable instructions stored in the computer-readable storage medium and execute the method described in the second aspect.

[0020] According to the fifth aspect of the present invention, there is provided a computer-readable storage medium, which is characterized in that the computer-readable storage medium stores computer instructions, and the computer instructions are used to cause the processor to execute the space station flight control task domain knowledge modeling and conversion method for HTN planning described in the first aspect, or the space station flight control task planning method based on HTN planning described in the second aspect.

[0021] Generally speaking, compared with the prior art, the above technical solution conceived by the present invention can achieve the following beneficial effects:

[0022] 1. The method for knowledge modeling and transformation in the space station flight control task field oriented to HTN planning provided by the present invention constructs a knowledge concept model of the space station flight control task field knowledge graph oriented to HTN planning and transforms it with the HTN planning field knowledge model; among them, the construction of the model ontology is to conceptually abstract the real environment based on the ontology modeling theory, define the core concepts, object properties and data properties of the space station flight control task field ontology, and edit and implement them through the protégé modeling tool, so as to construct a knowledge concept model of the space station flight control task field knowledge graph oriented to HTN planning. Further, a mapping relationship is established between the knowledge concept model of the space station flight control task field knowledge graph and the HTN planning field knowledge model, and a new specific domain knowledge model transformation method is proposed, which includes two parts: the Domain file transformation method and the Problem file transformation method, for obtaining relevant domain knowledge from the space station flight control task field knowledge graph and automatically transforming and generating the Domain file and Problem file required for HTN planning.

[0023] 2. The method for knowledge modeling and transformation in the space station flight control task field oriented to HTN planning provided by the present invention can reduce the difficulty of constructing the planning field model, improve the efficiency and effectiveness of model construction, provide effective domain knowledge model support for the development and execution of intelligent planning, and provide the necessary Domain file and Problem file input for the HTN planner. Since the construction of the domain knowledge graph combines the planning experience of flight control field experts and the actual flight control task planning process, it can be continuously updated and managed, and can guarantee the quality, completeness and effectiveness of domain knowledge. Therefore, this technology is applicable to the knowledge modeling and transformation in the actual space station flight control task planning field, so as to automatically generate the Domain file and Problem file required by the HTN planner, further improve the usability of the HTN planner, and provide support for the practical application of HTN intelligent planning technology. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 Schematic diagram of the HTN planning concept model;

[0025] Figure 2 Schematic diagram of the construction process of the space station flight control task field knowledge graph provided by the embodiment of the present invention;

[0026] Figure 3 Core concept map of the space station flight control task field knowledge defined by the pattern layer ontology provided by the embodiment of the present invention;

[0027] Figure 4 Schematic diagram of the mapping relationship between the domain knowledge graph concept model and the planning domain model provided by the embodiments of the present invention;

[0028] Figure 5 Schematic diagram of the conversion process of the domain file provided by the embodiments of the present invention;

[0029] Figure 6 Schematic diagram of the operator conversion process provided by the embodiments of the present invention;

[0030] Figure 7 Schematic diagram of the method conversion process provided by the embodiments of the present invention;

[0031] Figure 8 Schematic diagram of the conversion process of the problem file provided by the embodiments of the present invention. Detailed implementation manners

[0032] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0033] The embodiments of the present invention provide a method for domain knowledge modeling and conversion for space station flight control tasks oriented to HTN planning, including:

[0034] S1. According to the core concepts, object attributes and data attributes of the space station flight control task planning domain knowledge, create the model ontology of the domain knowledge graph, perform knowledge processing on the space station flight control task planning data to obtain the basic elements of the domain knowledge graph, and fill them into the model ontology to obtain the target model; wherein, the model includes a basic domain knowledge model and a planning domain knowledge model.

[0035] Preferably, in step S1, protégé is used to build the model ontology of the domain knowledge graph to obtain an initial OWL file; the target model obtained by filling the basic elements of the domain knowledge graph into the model ontology is a target OWL file.

[0036] Specifically, according to the core concepts, object attributes, and data attributes in the field of space station flight control mission planning, a model ontology of the domain knowledge graph is constructed using Protégé to obtain an initial OWL file. Among them, the model includes a basic domain knowledge model and a planning domain knowledge model. Knowledge processing is performed on the space station flight control mission planning data to obtain the basic elements of the domain knowledge graph and fill them into the initial OWL file to obtain a target OWL file. Among them, the basic elements include entities, relationships, and attributes.

[0037] Furthermore, Figure 2 Figure 1 shows four stages of the construction process of the space station flight control mission domain knowledge graph, namely domain knowledge analysis, schema layer ontology construction, data layer construction, and domain knowledge storage. The domain knowledge analysis stage aims to clarify the domain scope, sort out the knowledge content, and lay a foundation for the subsequent modeling process. With the support of flight control domain experts and ground control personnel, relevant materials, data, and expert experience of flight control mission planning are analyzed, and the key knowledge content and knowledge structure in the flight control mission planning field are sorted out and summarized, and formalized representation and description are carried out using a multi-tuple structure. The schema layer ontology construction stage is based on ontology modeling theory. According to the ontology construction principles and methods, the core concepts, object attributes, and data attributes in the space station flight control mission planning field are represented and modeled, and the Protégé modeling tool is used for ontology editing and implementation, and encoded representation is carried out using the OWL language, so as to complete the construction of the domain ontology concept model of the schema layer of the domain knowledge graph. The data layer construction stage is based on the constraints of the ontology concept model, and knowledge processing is performed on the simulated space station flight control mission planning experimental data, and three basic elements in the domain knowledge graph are extracted: entities, relationships, and attributes, and through knowledge fusion, specification integration and ambiguity elimination are carried out, and finally the triple structure of the domain knowledge is obtained to complete the entity filling of the data layer of the domain knowledge graph. The domain knowledge storage stage realizes the data storage of the domain knowledge graph by importing the encoded ontology OWL file into the Neo4j graph database, so as to support knowledge applications such as knowledge query and visualization, knowledge update and management, and knowledge model conversion of the domain knowledge graph.

[0038] S2. Establish mapping relationships between the Domain files of the planning domain knowledge model and the HTN planning domain model, and between the Problem files of the basic domain knowledge model and the HTN planning domain model, respectively.

[0039] Preferably, the basic domain knowledge model includes lunar events, flight control events, flight control instruction sequences, and resources.

[0040] The resources include TT&C resources, lighting resources, astronaut resources, electric power resources, and robotic arm resources.

[0041] The measurement and control resource category includes measurement and control modes and measurement and control entities.

[0042] Preferably, the planning domain knowledge model includes flight control tasks, flight control segments, domain rules, flight control actions, states, preconditions, and combined task sets;

[0043] The flight control tasks include flight control composite tasks and flight control atomic tasks;

[0044] The flight control composite tasks include lunar event tasks, flight control event tasks, send instruction sequence tasks, and select command - issuing time period tasks;

[0045] The flight control atomic tasks include merging measurement and control mode requirements, selecting command - issuing measurement stations, judging resource satisfaction, and occupying measurement and control command - issuing segments.

[0046] Preferably, the mapping relationship between the planning domain knowledge model and the Domain file of the HTN planning domain model is:

[0047] The flight control composite tasks and flight control atomic tasks respectively correspond to the composite tasks and atomic tasks in the HTN planning domain model;

[0048] The preconditions and combined task sets respectively correspond to the preconditions and sub - task lists in the HTN planning domain model;

[0049] The flight control segments correspond to the composite task decomposition method method in the HTN planning domain model;

[0050] The domain rules correspond to different decomposition strategies in the composite task decomposition method method in the HTN planning domain model;

[0051] The flight control actions correspond to the operators operator in the HTN planning domain model.

[0052] Preferably, the mapping relationship between the basic domain knowledge model and the Problem file of the HTN planning domain model is:

[0053] The hasName attribute and other attributes of each instance in the basic domain knowledge model respectively correspond to the state name and state parameter values in the HTN planning domain model;

[0054] The ID attribute of the lunar event instance in the basic domain knowledge model corresponds to the task parameter value in the HTN planning domain model.

[0055] Specifically, the core concepts and their hierarchical structures of the space station flight control task domain knowledge for HTN planning are as Figure 3 shown (that is Figure 2The core concepts of the knowledge in the space station flight control mission domain defined by the medium-mode layer ontology. Based on the analysis of the knowledge in the space station flight control mission planning domain, the knowledge in the space station flight control mission domain for HTN planning mainly consists of two-layer structures: one is the planning domain knowledge related to the flight control mission decomposition and planning process, such as the flight control hierarchical task network; the other is the basic domain knowledge provided to support the mission decomposition and planning, including various entities and their attributes in the real world. Therefore, the space station flight control mission domain ontology model is defined as two core classes: the space station basic domain ontology and the planning domain ontology.

[0056] Among them, the space station basic domain ontology is used to describe the basic knowledge that objectively exists and is externally input during the space station flight control mission planning process, usually serving as the basic data input for the space station flight control mission planning. It is formulated by different departments and specifically includes the knowledge description of three-layer structures: monthly event - flight control event - flight control instruction sequence, as well as the knowledge description of the required resources during the planning process. The core concepts mainly include four secondary subclasses: monthly event, flight control event, flight control instruction sequence, and resource. Among them, the resource class represents the general term for different resource types that can be provided in the real environment, including five tertiary subclasses: measurement and control resource, lighting resource, astronaut resource, electric power resource, and robotic arm resource. The measurement and control resource class can be further divided into two quaternary subclasses: measurement and control mode and measurement and control entity.

[0057] The space station planning domain ontology describes the planning knowledge such as domain rules and prerequisite states during the flight control mission planning process, involving specific flight control tasks and planning decomposition, handling rules, and completing the conversion of the knowledge model through the mapping relationship with the HTN planning elements. Based on the element expression of the HTN planning method, the core concepts of the planning knowledge described by the planning domain ontology mainly include seven secondary subclasses: task, flight control segment, domain rule, flight control action, state, prerequisite condition, and combined task set. The flight control task class includes two tertiary subclasses: flight control composite task and flight control atomic task. Among them, the flight control composite task class includes four quaternary subclasses: monthly event task class, flight control event task class, send instruction sequence task class, and select command sending time period task class; the flight control atomic task class includes four quaternary subclasses: merge measurement and control mode requirements class, select command sending measurement station class, judge resource satisfaction situation class, and occupy measurement and control command sending segment class.

[0058] The mapping relationship between the concept model of the domain knowledge graph and the planning domain model is as Figure 4As shown in the figure, for the domain ontology knowledge concept model structure of the space station flight control task domain knowledge graph, the planning domain knowledge model mainly corresponds to the decomposition structure and execution rules of the Domain file in the HTN planning domain model. The basic domain knowledge model is the basic source of states and corresponds to the initial state and task objectives in the Problem file of the HTN planning domain model. The tasks in the planning domain knowledge model include flight control composite tasks and flight control atomic tasks, corresponding to the composite tasks and atomic tasks in the HTN planning domain model respectively; flight control fragments are used to implement the flight control composite tasks to be executed, and they are mutually mapped with the composite task decomposition method method; domain rules are used to decompose flight control fragments. The same flight control fragment can be decomposed by multiple domain rules, which correspond to different decomposition strategies in the method decomposition method, that is, different branches; the domain rules include two parts: preconditions and combined task sets, corresponding to the preconditions and sub-task lists under the same branch of the method in the HTN planning domain model respectively. Flight control actions represent the execution of flight control atomic tasks and are mutually mapped with the operator operator to implement the instantiation operation of atomic tasks. States include a predicate set and a parameter set, which are derived from the basic domain knowledge model and can be converted into the addition effects and deletion effects after the execution of actions.

[0059] S3. Read the flight control action instances and their attributes and relationships with other instances in the planning domain knowledge model, and convert the flight control action instances into operators according to the mapping relationship; read the flight control fragment instances and their attributes and relationships with other instances in the planning domain knowledge model, and convert the flight control fragment instances into methods according to the mapping relationship; add the obtained operator set and method set to the Domain file to obtain the target Domain file.

[0060] S4. Read each instance and its attributes in the basic domain knowledge model respectively, and convert them into states according to the mapping relationship; read the monthly event instances and their attributes in the basic domain knowledge model, and convert them into tasks according to the mapping relationship; add the obtained state set and task set to the Problem file to obtain the target Problem file.

[0061] Specifically, the domain knowledge model conversion algorithm proposed by the present invention is based on the constructed space station flight control task domain knowledge graph. By using the OWL API interface, all relevant domain knowledge in the domain knowledge graph is read to obtain the classes, relationships, attributes, and instances in the domain ontology concept model, and then they are respectively converted into the Domain file and Problem file required for HTN planning.

[0062] Among them, the Domain file mainly converts the methods Method and operators Operator for describing the knowledge of the planning domain, and integrates and generates a domain knowledge file in PDDL format; the Problem file mainly converts the real-time state information of the current environment and the task objectives to be completed. The Domain file conversion algorithm includes three modules: the operator conversion module, the method conversion module, and the domain file conversion module. Among them, the operator conversion module converts the flight control action instances and their relationships and attributes into an operator set by reading them, the method conversion module converts the flight control fragment instances and their relationships and attributes into a method set by reading them, and the domain file conversion module creates relevant content according to the defined format of the domain file and adds the converted operator set and method set into it.

[0063] The conversion process of the domain file conversion module is as Figure 5 shown. The domain file conversion module is the core module of the Domain file conversion algorithm. The input of the conversion module is the OWL file of the domain knowledge of the space station flight control task constructed, and the output is the HTN planning Domain file in PDDL file format. The conversion module first creates the basic framework of the Domain file. For all instances of the knowledge model read, it respectively calls the operator conversion module and the method conversion module, adds the returned operator and method to the Domain file, and finally generates a complete Domain file.

[0064] The conversion process of the operator conversion module is as Figure 6 shown (i.e., the operator conversion process in Figure 4 ). The input of the operator conversion module is a flight control action instance A, and the output is a converted operator O. The conversion module obtains the <entity-relationship-entity> and <entity-attribute-attribute value> triple structure sets by reading the knowledge network related to the flight control action instance A, and then obtains the specific content required by the operator according to the object attributes and data attributes defined in the domain knowledge graph concept model, and finally combines to get a complete operator.

[0065] The conversion process of the method conversion module is as Figure 7 shown (i.e., Figure 4In the method conversion process), the input of the method conversion module is a flight control segment instance P, and the output is a decomposed method M that has been completed with conversion. The conversion module reads the knowledge network related to the flight control segment instance P, obtains its triple structure set, then respectively obtains the specific content required for the method according to the object attributes and data attributes defined in the domain knowledge graph concept model, and finally combines them to obtain a complete method. Since the lunar event task decomposition segment is a segment description that decomposes different lunar event tasks into different combinations of flight control event subtasks, the decomposition relationship comes from the inclusion relationship between lunar events and flight control events in the basic domain knowledge model and changes dynamically with different flight phases, while the decomposition rules for the corresponding flight control composite tasks described by other flight control segments are relatively fixed. Therefore, the method conversion module includes two parts: reading the lunar event task decomposition segment and other task flight control segments. Each part searches continuously along the node relationship according to the mapping relationship between the concept model and the planning model until the attribute value that meets the mapping relationship is obtained as the content filling of the method, and finally combines to obtain the final complete method.

[0066] The conversion process of the problem file conversion algorithm is as Figure 8 shown. The input of the algorithm is the OWL file of the space station flight control task domain knowledge constructed, and the output is the Problem file in the PDDL file format. The conversion algorithm first creates the basic framework of the Problem file, and respectively converts and constructs the state set and task set for all instances of the read knowledge model. For the state set, the name of the state is obtained through the hasName attribute, and the parameter values of other attributes are obtained for content filling; for the task set, the ID attribute of the lunar event instance is read and combined according to the expression form of the task. Finally, the converted state and task are added to the Problem file, and the complete Problem file is finally generated.

[0067] An embodiment of the present invention provides a space station flight control task planning method based on HTN planning. The target Domain file and the target Problem file obtained by using the method described in any of the above embodiments are input into the HTN planner to obtain a space station flight control task planning scheme.

[0068] An embodiment of the present invention provides a system for modeling and converting space station flight control task domain knowledge for HTN planning, including: a computer-readable storage medium and a processor;

[0069] The computer-readable storage medium is used to store executable instructions;

[0070] The processor is configured to read the executable instructions stored in the computer-readable storage medium and execute the knowledge modeling and transformation method for space station flight control mission planning as described in any of the above embodiments.

[0071] An embodiment of the present invention provides a space station flight control mission planning system based on HTN planning, including: a computer-readable storage medium and a processor;

[0072] The computer-readable storage medium is used to store executable instructions;

[0073] The processor is configured to read the executable instructions stored in the computer-readable storage medium and execute the space station flight control mission planning method based on HTN planning as described in the above embodiment.

[0074] An embodiment of the present invention provides a computer-readable storage medium storing computer instructions for causing a processor to execute the knowledge modeling and transformation method for the space station flight control mission field oriented to HTN planning as described in any of the above embodiments, or the space station flight control mission planning method based on HTN planning as described in the above embodiment.

[0075] Those skilled in the art can easily understand that the above are only preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for knowledge modeling and transformation of the space station flight control task domain oriented to HTN planning, characterized in that Including: S1. Create the model ontology of the domain knowledge graph according to the core concepts, object attributes, and data attributes of the domain knowledge in the space station flight control mission planning field, perform knowledge processing on the space station flight control mission planning data to obtain the basic elements of the domain knowledge graph, and fill them into the model ontology to obtain the target model; wherein, the model includes a basic domain knowledge model and a planning domain knowledge model; S2. Establish the mapping relationships between the planning domain knowledge model and the Domain file of the HTN planning domain model, and between the basic domain knowledge model and the Problem file of the HTN planning domain model respectively; S3. Read the flight control action instances and their attributes and relationships with other instances in the planning domain knowledge model, and convert the flight control action instances into operators according to the mapping relationship; read the flight control fragment instances and their attributes and relationships with other instances in the planning domain knowledge model, and convert the flight control fragment instances into methods according to the mapping relationship; add the obtained operator set and method set to the Domain file to obtain the target Domain file; S4. Read each instance and its attributes in the basic domain knowledge model respectively, and convert them into states according to the mapping relationship; read the monthly event instances and their attributes in the basic domain knowledge model, and convert them into tasks according to the mapping relationship; add the obtained state set and task set to the Problem file to obtain the target Problem file.

2. The method according to claim 1, wherein The basic domain knowledge model includes monthly events, flight control events, flight control instruction sequences, and resources; The resources include TT&C resources, illumination resources, astronaut resources, electric power resources, and robotic arm resources; The TT&C resource class includes TT&C modes and TT&C entities.

3. The method according to claim 1 or 2, characterized in that, The planning domain knowledge model includes flight control tasks, flight control fragments, domain rules, flight control actions, states, preconditions, and combined task sets; The flight control tasks include flight control composite tasks and flight control atomic tasks; The flight control composite tasks include monthly event tasks, flight control event tasks, send instruction sequence tasks, and select command-issuing time period tasks; The flight control atomic tasks include merge TT&C mode requirements, select command-issuing TT&C stations, judge resource satisfaction, and occupy TT&C command-issuing fragments.

4. The method according to claim 3, wherein The mapping relationship between the planning domain knowledge model and the Domain file of the HTN planning domain model is: The flight control composite tasks and flight control atomic tasks respectively correspond to the composite tasks and atomic tasks in the HTN planning domain model; The preconditions and combined task sets respectively correspond to the preconditions and sub-task lists in the HTN planning domain model; The flight control fragments correspond to the composite task decomposition method method in the HTN planning domain model; The domain rules correspond to different decomposition strategies in the composite task decomposition method method in the HTN planning domain model; The flight control actions correspond to the operators operator in the HTN planning domain model.

5. The method according to claim 3 or 4, characterized in that, The mapping relationship between the basic domain knowledge model and the Problem file of the HTN planning domain model is: The hasName attribute and other attributes of each instance in the basic domain knowledge model respectively correspond to the state name and state parameter values in the HTN planning domain model; The ID attribute of the monthly event instance in the basic domain knowledge model corresponds to the task parameter value in the HTN planning domain model.

6. The method according to claim 1, wherein In step S1, the model ontology of the domain knowledge graph is constructed using protégé to obtain an initial OWL file; the target model obtained by filling the basic elements of the domain knowledge graph into the model ontology is the target OWL file.

7. A space station flight control mission planning method based on HTN planning, characterized in that, The target Domain file and target Problem file obtained by using the method according to any one of claims 1-6 are input into the HTN planner to obtain a space station flight control task planning scheme.

8. A knowledge modeling and transformation system for flight control task planning in a space station, characterized in that, Including: A computer-readable storage medium and a processor; The computer-readable storage medium is used to store executable instructions; The processor is used to read the executable instructions stored in the computer-readable storage medium and execute the method according to any one of claims 1-6.

9. A space station flight control mission planning system based on HTN planning, characterized in that, Including: A computer-readable storage medium and a processor; The computer-readable storage medium is used to store executable instructions; The processor is used to read the executable instructions stored in the computer-readable storage medium and execute the method according to claim 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions, and the computer instructions are used to cause the processor to execute the method for space station flight control task domain knowledge modeling and transformation for HTN planning according to any one of claims 1-6, or the method for space station flight control task planning based on HTN planning according to claim 7.

Images