An unmanned aerial vehicle space-air-ground coordination method

Abstract

This invention relates to the field of air-space-ground collaborative technology, and particularly to a method for UAV air-space-ground collaborative technology, comprising: constructing a collaborative key area state in the air-space-ground collaborative layer; writing collaborative information from air-space-ground collaborative participants into the collaborative key area state according to a preset attribute classification method, and handling conflicts between multiple source update requests for the same state attribute; dynamically determining the collaborative decision-making relationship based on the collaborative key area state characterizing the state uncertainty of air-space-ground collaborative participants within the key area, and generating a collaborative responsibility state; and identifying unobservable collaborative areas by combining the available observation information of collaborative participants, inferring and updating their states, generating and issuing collaborative tasks for airborne UAVs. This method ensures that the collaborative area state no longer depends on the data integrity of any single collaborative participant, thus enabling unified maintenance of the collaborative area state even when multiple sources update at different frequencies or when local information is missing.

Description

Technical Field

[0001] This invention relates to the field of air-space-ground coordination technology, and in particular to a method for air-space-ground coordination of unmanned aerial vehicles (UAVs). Background Technology

[0002] In emergency applications such as forest fire monitoring and response, a collaborative air-space-ground technology system has gradually formed, involving space-based systems, airborne drones, and ground-based command systems. Space-based systems are typically used to acquire information on the distribution and overall evolution of fires over large areas; airborne drones are used to acquire high-resolution real-time sensing information for localized areas; and ground-based command systems are used to integrate multi-source information and generate dispatch and response decisions. Existing drone-based air-space-ground collaborative methods generally aggregate information from different collaborative participants through multi-source data fusion or centralized situational awareness displays to support fire analysis and collaborative decision-making.

[0003] In the actual operation of rapid forest fire evolution, there are significant differences in the time scale, coverage and update stability of information obtained by different air-space-ground collaborative participants. Existing collaborative methods usually rely on a certain type of data source as the main basis for collaborative status. When the update of this data source is limited or the information is missing, it is difficult to maintain the status of the collaborative area continuously and uniformly, which in turn affects the continuity of air-space-ground collaborative operation. Summary of the Invention

[0004] To overcome the above shortcomings, this invention provides a method for UAV air-ground-space collaboration, which aims to improve the problem that when the update of a certain data source is limited or information is missing, it is difficult to maintain the status of the collaborative area continuously and uniformly, thus affecting the continuity of air-ground-space collaborative operation.

[0005] This invention provides the following technical solution: a method for UAV air-ground collaborative operation, comprising the following steps: S1. Construct the state of key collaborative areas in the space-air-ground collaborative layer, making them independent of any single data source; S2. Write the collaboration-related information from the participants in the air-space-ground collaboration into the corresponding status attributes of the collaborative key area status according to the preset attribute classification method, and handle the conflict of update requests from different sources for the same status attribute, and write the processing results into the collaborative key area status. S3. Based on the state of the key collaborative area, characterize the state uncertainty of the participants in the air-space-ground collaboration within the key collaborative area, and determine the responsibility relationship for collaborative decision-making in the air-space-ground collaboration process based on the state uncertainty, thereby generating the collaborative responsibility state. S4. Determine the collaborative decision-making entity based on the collaborative responsibility status, and generate collaborative decision input based on the status information with low uncertainty in the collaborative key area status; S5. Based on the available observation information of the participants in the air-space-ground collaboration, identify areas that cannot be effectively observed and mark them as unobservable collaborative areas. Combine the adjacent state information and historical state records of the key collaborative areas to infer and update the state of the unobservable collaborative areas. S6. Based on the status of key collaborative areas, collaborative responsibility status, status of unobservable collaborative areas, and collaborative decision inputs, generate collaborative tasks for airborne UAVs and issue them for execution.

[0006] By adopting the above technical solution, a collaborative key area state independent of any single data source is constructed in the air-space-ground collaboration layer. Collaboration-related information from different air-space-ground collaboration participants is written into this collaborative key area state in a attribute-based manner, thus freeing the collaborative area state from continuous updates dependent on a single type of data source. When information updates from some air-space-ground collaboration participants are restricted or temporarily unavailable, the collaborative key area state can still continuously evolve based on valid information from other participants and existing state content. This ensures the continuous and unified maintenance of the collaborative area state during the air-space-ground collaboration process, preventing interruptions to air-space-ground collaboration operations due to limitations of a single data source.

[0007] Preferably, in S1, the construction of the collaborative key region state includes: In the space-air-ground collaborative layer, identify key areas as objects of collaborative concern; Establish corresponding state description structures for key areas to characterize the state attributes of key areas in the air-space-ground coordination process; By decoupling the state description structure from the participants in the space-air-ground collaboration, the state of the key collaborative areas can exist independently of any single data source and be jointly updated by the participants in the space-air-ground collaboration.

[0008] Preferably, in S2, the step of writing the corresponding state attributes of the collaborative key area state according to the preset attribute division method includes: The relevant information from the participants in the air-space-ground collaboration is classified by attribute type, and the relevant information is written into the mutually isolated state attribute set in the state of the key collaborative area according to the classification result; Different state attribute sets are used to characterize the state changes of key areas under different collaborative dimensions, and the collaborative information written into different state attribute sets does not participate in the same update competition in subsequent state update processes.

[0009] Preferably, in S2, the conflict handling of update requests from different sources for the same state attribute includes: Within the same state update cycle, detect whether multiple update requests for the same state attribute exist simultaneously; When multiple update requests are detected, the priority of the multiple update requests is determined by combining the collaborative responsibility status; Based on the priority determination results, full writes are performed on update requests with higher priority, while restricted writes or delayed writes are performed on update requests with lower priority.

[0010] Preferably, in S3, the characterization of the state uncertainty of the participants in the air-space-ground collaboration within the key collaborative area includes: Within a preset state evaluation time window, the update frequency of each state attribute is statistically analyzed to characterize the degree of uncertainty in the changes of state attributes. Calculate the magnitude of change of each state attribute within the state evaluation time window to characterize the degree of uncertainty of the state attribute changes; The correlation changes among different state attributes within the state evaluation time window are analyzed to determine the interaction relationships between state attributes, which are used to characterize the joint uncertainty among state attributes.

[0011] Preferably, in S3, the determination of the collaborative decision-making relationship in the air-space-ground coordination process includes: Based on state uncertainty, calculate the information entropy of the participants in the air-space-ground collaboration within the key collaborative area; Compare the information entropy of different participants in the air-space-ground collaboration and select the participant with the lower information entropy as the decision-maker for the current collaboration stage. When the information entropy changes and the preset collaborative adjustment conditions are met, the collaborative decision-making relationship is adjusted and the collaborative responsibility status is updated.

[0012] Preferably, in S4, generating collaborative decision input based on state information with low uncertainty in the collaborative key area state includes: Based on the collaborative responsibility status, select the status attributes corresponding to the current collaborative decision-making relationship from the status of key collaborative areas; The selected state attributes are reorganized to generate a decision input structure that meets the processing requirements of the collaborative decision generation subject; The decision input structure is provided as collaborative decision input to the collaborative decision generation entity.

[0013] Preferably, in S5, determining and marking the regions where effective observations cannot be formed as unobservable cooperative regions includes: During the collaborative operation, observation availability information for the collaborative area is obtained from each of the participants in the air-space-ground collaboration. Based on observation availability information, determine whether the collaborative area continues to lack effective observations from space-air-ground collaborative participants during the preset observation evaluation period. When a cooperative region is determined to have a persistent lack of effective observations, it is marked as an unobservable cooperative region.

[0014] Preferably, in S5, the step of inferring and updating the state of unobservable cooperative regions by combining the adjacent state information and historical state records of the cooperative key regions includes: Obtain the status of key collaborative regions that are spatially or in collaborative relationships with unobservable collaborative regions; Obtain historical state records of the unobservable cooperative region before it enters an unobservable state; Based on the correlation between the state of adjacent collaborative key areas and historical state records, the current state of unobservable collaborative areas is inferred and updated.

[0015] Preferably, in S6, the cooperative task of generating the airborne UAV includes: Based on the collaborative responsibility status, determine the collaborative role that the airborne UAV plays in the current collaborative phase; By combining the status of key collaborative areas and the status of unobservable collaborative areas, the target areas corresponding to collaborative roles are determined from the key collaborative areas. Based on the state characteristics of the target area, generate airborne UAV collaborative missions that match the collaborative roles.

[0016] The present invention has the following beneficial effects: 1. In this invention, by constructing a collaborative key area state independent of any single data source in the air-space-ground collaborative layer, and allowing collaborative information from different air-space-ground collaborative participants to be written and updated according to the attribute division method, the collaborative area state no longer depends on the data integrity of a certain collaborative participant. Thus, even when the update frequency of multi-source information is different or local information is missing, the collaborative area state can still be uniformly maintained.

[0017] 2. In this invention, the uncertainty of the state of participants in the air-space-ground collaboration within the key collaborative area is characterized based on the state of the key collaborative area. Furthermore, information entropy is calculated based on the state uncertainty to determine the responsibility for collaborative decision-making. This enables the collaborative decision-making entity to make dynamic adjustments according to changes in the degree of mastery of the collaborative area's state, thereby avoiding the continuous assumption of collaborative decision-making responsibility by collaborative participants with high state uncertainty during collaborative operation.

[0018] 3. In this invention, by identifying collaborative regions that cannot be effectively observed during collaborative operation, and by combining the status of adjacent key collaborative regions and historical status records, the status of unobservable collaborative regions is inferred and updated, so that unobservable regions can still participate in collaborative decision-making and collaborative task generation processes even in the absence of real-time observation information, thereby avoiding the interruption of collaborative region status due to the lack of local observations during collaborative operation. Attached Figure Description

[0019] Figure 1 This is a flowchart of a UAV air-ground collaborative method proposed in this invention. Detailed Implementation

[0020] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] In a first embodiment of the present invention, the present invention provides a method for UAV air-ground collaborative operation, such as... Figure 1 As shown, it includes the following steps: S1. Construct the state of key collaborative areas in the space-air-ground collaborative layer, making them independent of any single data source; Furthermore, in S1, constructing the collaborative key region state includes: In the space-air-ground collaborative layer, identify key areas as objects of collaborative concern; Establish corresponding state description structures for key areas to characterize the state attributes of key areas in the air-space-ground coordination process; By decoupling the state description structure from the participants in the space-air-ground collaboration, the state of the key collaborative areas can exist independently of any single data source and be jointly updated by the participants in the space-air-ground collaboration.

[0022] Specifically, the air-space-ground coordination layer serves as a unified logical layer for state interaction and collaborative operation among air-based, space-based, and ground-based collaborative participants. It is used to carry key state information related to collaborative operation. The state of key collaborative areas is constructed in this air-space-ground coordination layer, so that the state of key collaborative areas exists independently of the data source of any single collaborative participant, thereby providing a unified state carrying foundation for multi-source collaborative state updates and subsequent collaborative decisions. In the specific implementation process, when constructing the state of key collaborative areas, the key areas that are the objects of collaborative attention are first determined in the air-space-ground collaborative layer. The key areas are used to represent the spatial areas or task areas that need to be analyzed or processed in the current collaborative task or collaborative operation phase. The key areas can be described by spatial range, time range or task identifier, for example, in the form of a set of spatial coordinates or a spatial region function. After identifying the key areas, a corresponding state description structure is established for each key area. This structure is used to uniformly model the state attributes of the key areas during the air-space-ground coordination process. The state description structure can be represented as a multi-dimensional state vector, which carries the overall state information of the key areas during the coordinated operation. For example, the state of the coordinated key area can be represented as: ; in, Indicates at time Next The overall status of key collaborative areas Indicates the key collaborative area in the first State values ​​on each state attribute dimension This indicates the number of dimensions for the state attribute. The number of dimensions and their specific meanings are configured according to the needs of the collaborative application. The above state attributes It can be used to characterize the state changes of key areas under different collaborative dimensions. Different state attributes are stored independently but logically related in the state description structure. Through the above vectorization method, the state description structure enables the multi-source and multi-dimensional state of key areas to exist in the air-space-ground collaborative layer in a unified data structure form, thereby providing a foundation for subsequent state updates, state analysis and collaborative decision-making. To ensure that the state of the key collaborative area is not dependent on any single collaborative participant, the state description structure is decoupled from the space-air-ground collaborative participants. That is, the state of the key collaborative area is uniformly maintained by the space-air-ground collaborative layer, and each collaborative participant only participates in the updating and use of the state of the key collaborative area through the state writing interface or the state access interface, without directly controlling the definition method or lifecycle of the state description structure. Through the above decoupling method, collaboration-related information from different participants in the space-air-ground collaboration is mapped to the corresponding state attributes in the state of the key collaborative areas. This enables the state of key collaborative areas to persist and evolve dynamically within the space-air-ground collaborative layer. This state of key collaborative areas serves as the foundational state object for characterizing subsequent state uncertainties, determining collaborative decision-making relationships, and generating collaborative tasks. It is continuously maintained and updated throughout the entire space-air-ground collaborative operation process.

[0023] S2. Write the collaboration-related information from the participants in the air-space-ground collaboration into the corresponding status attributes of the collaborative key area status according to the preset attribute classification method, and handle the conflict of update requests from different sources for the same status attribute, and write the processing results into the collaborative key area status. Furthermore, in S2, the corresponding state attributes of the collaborative key area state are written according to the preset attribute division method, including: The relevant information from the participants in the air-space-ground collaboration is classified by attribute type, and the relevant information is written into the mutually isolated state attribute set in the state of the key collaborative area according to the classification result; Different state attribute sets are used to characterize the state changes of key areas under different collaborative dimensions, and the collaborative information written into different state attribute sets does not participate in the same update competition in subsequent state update processes.

[0024] Furthermore, in S2, conflict handling for update requests from different sources targeting the same state attribute includes: Within the same state update cycle, detect whether multiple update requests for the same state attribute exist simultaneously; When multiple update requests are detected, the priority of the multiple update requests is determined by combining the collaborative responsibility status; Based on the priority determination results, full writes are performed on update requests with higher priority, while restricted writes or delayed writes are performed on update requests with lower priority.

[0025] Specifically, during the collaborative operation, the participants in the air-space-ground collaboration continuously generate collaborative-related information related to the key collaborative area. The air-space-ground collaboration layer receives the above-mentioned collaborative-related information and writes it into the corresponding status attribute of the collaborative key area status according to the preset attribute division method. At the same time, during the writing process, conflict handling is performed on update requests from different sources for the same status attribute, and the result of conflict handling is written into the collaborative key area status so that the collaborative key area status can be consistently referenced by multiple parties within the air-space-ground collaboration layer. In the specific implementation process, collaboration-related information is normalized and encapsulated by the space-air-ground collaboration layer before entering the writing process. This allows data from different collaboration participants to participate in attribute partitioning and conflict resolution in a unified data item format. The data item for collaboration-related information can be represented as: ; in, This indicates that candidate data items are written at once. This indicates the source identifier for collaborative information, used to distinguish different participants in the space-air-ground collaboration. Indicates key collaborative area identifiers. This indicates the timestamp corresponding to the data item. This represents the content vector or set of content fields for the data item. The dimensions and field types of the content vector are configured according to the requirements of the collaborative application. The source identifier, region identifier, and time stamp are used to establish write associations and time sequence relationships within the collaborative layer. The preset attribute partitioning method is used to map collaboration-related information to different state attribute sets in the collaborative key area state. The attribute partitioning method can be implemented in the form of an attribute type discrimination function, for example, defining an attribute type discrimination function: ; in, This represents an attribute type discrimination function, whose input is a content vector or a set of content fields related to collaboration. The output is an attribute type identifier. The attribute type identifier is used to indicate the category of the status attribute set to which the collaboration-related information should be written. The set of values ​​for the attribute type identifier can be represented as follows: , This represents a pre-defined set of attribute types, which is used to cover the state change expression requirements of key collaborative areas under different collaborative dimensions. Based on the attribute type discrimination result, collaboration-related information is written into mutually isolated state attribute sets in the state of key collaboration areas. These state attribute sets are used to isolate and store state information from different collaboration dimensions at the data structure level, ensuring that state updates under different collaboration dimensions do not compete for the same update in subsequent processing. The state attribute sets can optionally be represented as follows: ,in Indicates key areas In attribute type The isolated storage of the set of state attributes can be achieved by establishing independent storage areas, independent cache queues or independent state register structures for different attribute types, so that the collaborative related information written to different set of state attributes will not be merged into the same conflict decision domain during the update cycle. After determining the attribute type and identifying the target state attribute set, the space-air-ground coordination layer performs conflict detection on update requests for the same state attribute. The state update cycle is used to define the temporal boundaries of conflict detection and conflict handling. The state update cycle can be implemented using a fixed duration, event triggering, or sliding window method. Within the same state update cycle, the space-air-ground coordination layer aggregates the set of update requests with the same target. The set of update requests can be represented as: ; in, This indicates that the key areas are targeted within the current state update cycle. The A set of update requests for each state attribute. This represents a state attribute locating function, used to further locate a specific state attribute identifier within a defined set of attribute types. ,when This indicates that multiple update requests are simultaneously targeting the same state attribute, triggering the conflict resolution process. When multiple update requests are detected simultaneously, the space-air-ground coordination layer prioritizes these requests based on the coordination responsibility status. The coordination responsibility status characterizes which participant is responsible for generating collaborative decisions in the current coordination phase. The coordination responsibility status can optionally be represented as... ,in For collaborative responsibility status identification, , This represents the set of identifiers for collaborative participants. The priority determination criterion is used to associate the collaborative responsibility status with the identifier of the update request source. The priority determination function can optionally be expressed as: ; in, This represents the priority determination function, whose input is the update request data item. , Indicates a high priority identifier. Indicates a low priority identifier. This indicates that the source identifier of the update request is being updated. The above determination method enables the collaborative responsibility status to participate in conflict resolution, thereby providing an executable write control basis when there are multiple concurrent writes to the same status attribute. Based on the priority determination result, a full write is performed on update requests with higher priority. A full write involves writing the content vector or set of content fields of the update request into the target state attribute according to a predetermined mapping rule and updating its timestamp, source identifier, and other metadata. Optionally, a full write can be represented as updating the target state attribute to... ,in This represents the state write operation function, which is used to map the content in the update request data item to the state attribute value space and form a new state attribute value. For update requests with lower priority, restricted write or delayed write is performed. Restricted write is used to retain some information of the low-priority update request without directly overwriting the target state attribute value in the current period. The form of restricted write can be to write only the confidence flag, only the difference, only the candidate value set, or only the source and timestamp, etc. Delayed write is used to temporarily store the low-priority update request to the next state update period or a preset cache queue for further processing. The choice between restricted write or delayed write can be configured according to the needs of collaborative applications, such as the relationship between the timestamp of the update request, the magnitude of the change in the content vector, or the system load status, but there is no limitation on the specific selection method. Through the aforementioned attribute partitioning and conflict handling process, the space-air-ground collaboration layer can write collaboration-related information from different collaboration participants into the corresponding status attributes of the collaborative key area status. When concurrent updates occur in the same status attribute, executable priority control and write control are formed, ensuring that the collaborative key area status maintains a continuous and traceable status update trajectory during collaborative operation. This provides a stable data foundation for subsequent characterization of status uncertainty based on the collaborative key area status and determination of collaborative decision-making relationships.

[0026] S3. Based on the state of the key collaborative area, characterize the state uncertainty of the participants in the air-space-ground collaboration within the key collaborative area, and determine the responsibility relationship for collaborative decision-making in the air-space-ground collaboration process based on the state uncertainty, thereby generating the collaborative responsibility state. Furthermore, in S3, the uncertainty in the state of the participants in the air-space-ground collaboration within the key collaborative area includes: Within a preset state evaluation time window, the update frequency of each state attribute is statistically analyzed to characterize the degree of uncertainty in the changes of state attributes. Calculate the magnitude of change of each state attribute within the state evaluation time window to characterize the degree of uncertainty of the state attribute changes; Analyze the correlation changes among different state attributes within the state evaluation time window to determine the interaction relationships between state attributes, which is used to characterize the joint uncertainty among state attributes.

[0027] Furthermore, in S3, determining the roles of those responsible for collaborative decision-making during the air-space-ground coordination process includes: Based on state uncertainty, calculate the information entropy of the participants in the air-space-ground collaboration within the key collaborative area; Compare the information entropy of different participants in the air-space-ground collaboration and select the participant with the lower information entropy as the decision-maker for the current collaboration stage. When the information entropy changes and the preset collaborative adjustment conditions are met, the collaborative decision-making relationship is adjusted and the collaborative responsibility status is updated.

[0028] Specifically, the air-space-ground coordination layer characterizes the uncertainty of the state of the air-space-ground coordination participants in the key coordination area based on the state of the key coordination area, and determines the responsibility relationship of the coordination decision in the air-space-ground coordination process according to the state uncertainty to generate the coordination responsibility state. The coordination responsibility state is used to indicate the identifier of the coordination participant who assumes the responsibility for generating coordination decision in the current coordination stage. In the specific implementation process, the representation of state uncertainty is carried out within a preset state evaluation time window, which is denoted as: ,in Indicates the start time of the window. Indicates the duration of the window and Key areas are marked as The collaborative participants are identified as follows: The status attribute identifier is denoted as ,in Taken from the set of collaborative participants , Taken from key region set , Taken from the set of state attribute indexes , This indicates the number of state attribute dimensions contained in the key area state description structure. The update records of the state attributes within the time window come from the time-series storage formed by the write results of the collaborative key area state. To characterize the uncertainty of state attribute changes, the space-air-ground coordination layer statistically analyzes the update frequency of each state attribute within the state evaluation time window, targeting the coordination participants. Key areas and state attributes The set of valid update events generated by the participating policy for the state attribute within the time window is denoted as... A valid update event refers to an event that the participant updates to a state attribute. Write requests, after conflict resolution, are accepted and written into the update event of the collaborative critical area state, with the event set cardinality... This indicates the number of valid updates, and the update frequency is defined as: ; in Indicates the participating parties In key areas Internal state attributes The activity level is updated per unit of time; To further characterize the uncertainty of state attribute changes, the space-air-ground coordination layer calculates the magnitude of change of each state attribute within the state assessment time window, targeting the coordination participants. Key areas and state attributes The sequence of state attribute values ​​written into the collaborative key area state within the time window and originating from that participant is denoted as... ,in Indicates the first The timestamp corresponding to the first valid write and satisfying , This indicates the status attribute at that timestamp. The value of , This indicates the number of valid updates to the status attribute within the time window, and defines the magnitude of the change as follows: ; in Indicates the participating parties In key areas Internal state attributes The intensity of the value fluctuation; To characterize the joint uncertainty among state attributes, the space-air-ground coordination layer analyzes the correlation changes of different state attributes within the state assessment time window to determine the interaction relationships between state attributes. Correlation change analysis can be achieved through statistical correlation indicators, such as correlation coefficients, mutual information, or other indicators reflecting the degree of synchronous change, targeting the participating parties. Key areas and arbitrary state attribute pairs and Within a time window, the correlation degree formed based on their respective value sequences is denoted as... ,in Used to characterize state attributes With state attributes Based on the combined change intensity, the space-air-ground collaborative layer can construct a set of interaction relationships based on the correlation degree of all attribute pairs. , as a representation input for joint uncertainty; The air-space-ground coordination layer is based on the update frequency. Variation range and the set of interaction relationships Forming Participants In key areas The uncertainty of the state within the region is characterized, and the information entropy of the participants in the key region is calculated based on the uncertainty of the state. The information entropy is used to quantify the degree of uncertainty of the participants' state expression. The calculation of information entropy is based on the discrete probability distribution formed by the state attributes within the time window, targeting the participants. Key areas and state attributes Construct a set of discrete states ,in Indicates the first A discrete state interval or discrete state label. Represents the number of discrete states and The construction of a discrete state set can be achieved by dividing the value interval equally, dividing it by quantile points, or dividing it by a preset state level, and by counting the number of times the state attribute value falls into each discrete state. ,in Indicates the state attribute within the time window. The effective values ​​fall into discrete states The number of times, and from this, the probability distribution is obtained: ; in Represents state attributes Falling into discrete state The probability, the probability distribution satisfies ; After obtaining the probability distribution, the space-air-ground collaborative layer calculates the information entropy of the state attributes: ; in Indicates the participating parties In key areas Internal state properties Information entropy This represents a logarithmic operation, and the logarithmic base remains consistent throughout the same implementation. The logarithmic base can be either 2 or the natural logarithm. As one possible approach; When it is necessary to integrate the information entropy of multiple state attributes, the space-air-ground collaborative layer calculates the integrated information entropy of the participants in the key area: ; in Indicates the participating parties In key areas The comprehensive information entropy within, Indicates key areas Internal state properties The weighting coefficients are used to characterize the relative importance of different state attributes in cooperative operation, and the weighting coefficients satisfy the following conditions: and ; The air-space-ground coordination layer compares the different collaborative participants in key areas. Comprehensive information entropy within Based on the size of the information entropy, the collaborating participant with the lower information entropy is selected as the collaborative decision-maker for the current collaborative stage, and the identifier of this participant is written into the collaborative responsibility status. The collaborative responsibility status can be represented as... ,in And used to indicate key areas The party responsible for collaborative decision-making at the current collaborative stage; During continuous collaborative operation, the space-air-ground collaborative layer updates the information entropy periodically or through event-triggered updates. When the information entropy changes and preset collaborative adjustment conditions are met, the collaborative decision-making relationship is adjusted and the collaborative responsibility status is updated. The preset collaborative adjustment conditions are used to limit the timing of updating the collaborative responsibility status. The preset collaborative adjustment conditions may include one or more conditions such as changes in the ranking of comprehensive information entropy, the change in comprehensive information entropy reaching a threshold, the rate of change in comprehensive information entropy reaching a threshold, or a collaborative phase switching trigger, so that the collaborative responsibility status can dynamically change with the evolution of the status of key collaborative areas.

[0029] S4. Determine the collaborative decision-making entity based on the collaborative responsibility status, and generate collaborative decision input based on the status information with low uncertainty in the collaborative key area status; Furthermore, in S4, generating collaborative decision inputs based on state information with low uncertainty in the collaborative key area states includes: Based on the collaborative responsibility status, select the status attributes corresponding to the current collaborative decision-making relationship from the status of key collaborative areas; The selected state attributes are reorganized to generate a decision input structure that meets the processing requirements of the collaborative decision generation subject; The decision input structure is provided as collaborative decision input to the collaborative decision generation entity.

[0030] Specifically, after obtaining the collaborative responsibility status, the air-space-ground collaborative layer determines the collaborative decision-making entity for the current collaborative stage based on the collaborative responsibility status, and generates collaborative decision input based on the state information with low uncertainty in the state of the key collaborative area, so that the collaborative decision-making entity can execute the collaborative decision-making process based on the relatively stable state information in the current collaborative stage. In practical implementation, the collaborative responsibility status is used to indicate the identifier of the collaborative participant who assumes responsibility for generating collaborative decisions in the current collaborative phase. The collaborative responsibility status can be represented as follows: ,in The key area is marked as The key collaborative areas correspond to the collaborative decision-making entities at the current collaborative stage, and these entities are drawn from the set of participants in the space-air-ground collaborative process. The space-air-ground coordination layer assigns the collaborative decision generation process to the corresponding collaborative participants based on the collaborative responsibility status. To generate collaborative decision inputs based on state information with low uncertainty in the state of key collaborative areas, the space-air-ground collaborative layer first selects a set of state attributes from the state of key collaborative areas that corresponds to the current collaborative decision-making relationship, based on the collaborative responsibility state. The set of state attributes is used to represent a subset of state attributes provided or dominated by the current collaborative decision-making entity. The selection of the subset of state attributes is based on the correspondence between the collaborative responsibility state and the source of the state attributes, so that the selected state attributes have relatively low uncertainty in the current collaborative stage. The uncertainty of state attributes originates from the representation results of state uncertainty in the state of key collaborative areas. The air-space-ground collaborative layer sorts or classifies the uncertainty at the state attribute level, and on this basis, selects state attributes with uncertainty below a preset threshold as a candidate state attribute set. The uncertainty threshold can be set according to the collaborative task type, collaborative stage characteristics, or decision-making timeliness requirements. The candidate state attribute set is used as the source of basic state information for collaborative decision input. After selecting the state attributes, the space-air-ground coordination layer reorganizes the selected state attributes to generate a decision input structure that meets the processing requirements of the collaborative decision generation subject. The decision input structure is used to organize the input information that the collaborative decision generation subject can directly use in a structured manner. The decision input structure can be represented as follows: ; in Indicates key areas The collaborative decision-making input structure, Indicates key areas The state attribute is identified as The value of the state attribute, This represents the set of indexes of the selected low-uncertainty state attributes. This represents a state attribute reorganization mapping function, used to combine multiple state attribute values ​​into a unified decision input structure according to preset organization rules; State attribute recombination mapping function The state attribute reorganization mapping function is used to unify the state attributes in terms of data format, dimensional structure and semantic organization. It can include processing steps such as standardization, dimension alignment, format encapsulation or semantic mapping of state attributes so that the generated decision input structure can match the internal decision model or decision logic of the collaborative decision generation subject. The specific implementation of the state attribute reorganization mapping function can be configured according to the type of collaborative decision generation subject. After generating the decision input structure, the space-air-ground coordination layer provides the decision input structure as the collaborative decision input to the collaborative decision generation subject, enabling the collaborative decision generation subject to execute the collaborative decision generation process based on the state information with low uncertainty in the current collaborative stage. The collaborative decision input is used as the basic input for generating collaborative tasks, adjusting collaborative strategies, or updating collaborative control parameters in subsequent collaborative operations.

[0031] S5. Based on the available observation information of the participants in the air-space-ground collaboration, identify areas that cannot be effectively observed and mark them as unobservable collaborative areas. Combine the adjacent state information and historical state records of the key collaborative areas to infer and update the state of the unobservable collaborative areas. Furthermore, in S5, areas where effective observations cannot be formed and are marked as unobservable cooperative regions include: During the collaborative operation, observation availability information for the collaborative area is obtained from each of the participants in the air-space-ground collaboration. Based on observation availability information, determine whether the collaborative area continues to lack effective observations from space-air-ground collaborative participants during the preset observation evaluation period. When a cooperative region is determined to have a persistent lack of effective observations, it is marked as an unobservable cooperative region.

[0032] Furthermore, in S5, by combining the adjacent state information and historical state records of the collaborative key region state, the state of the unobservable collaborative region is inferred and updated, including: Obtain the status of key collaborative regions that are spatially or in collaborative relationships with unobservable collaborative regions; Obtain historical state records of the unobservable cooperative region before it enters an unobservable state; Based on the correlation between the state of adjacent collaborative key areas and historical state records, the current state of unobservable collaborative areas is inferred and updated.

[0033] Specifically, the air-space-ground coordination layer assesses the observability of the coordination area based on the available observation information of the air-space-ground coordination participants. When it is detected that a coordination area cannot form effective observations, the area is marked as an unobservable coordination area. At the same time, the state of the unobservable coordination area is inferred and updated by combining the adjacent state information and historical state records of the key coordination area, so as to maintain the continuity of the state of the key coordination area during the coordination operation. In the specific implementation process, the space-air-ground coordination layer acquires observation availability information for the coordination area from each participating party during the coordination operation. This observation availability information characterizes whether the participating parties can effectively observe the coordination area within the current coordination phase. The coordination area is identified as [identified by the relevant identifier]. The collaborative participants are identified as follows: Observation availability information can be represented as a binary or continuous index. As one possible approach, observation availability can be represented as: ; in Indicates collaborative participants At any moment For collaborative regions The observation availability status; The space-air-ground integrated layer during the preset observation and evaluation period The system performs statistical analysis on observation availability information, including... Indicates the start time of the observation and evaluation period. Indicates the duration of the observation and evaluation period, for the collaborative region. The observation availability of all collaborating participants within this time period is aggregated, and the regional observability is defined as: ; in This represents the set of participants in the space-air-ground collaboration. Indicates collaborative region The overall observability during the observation and evaluation period is used to reflect the overall observation coverage of the collaborative region in terms of time and participants. The space-air-ground synergy layer will improve regional observability Compared with the preset observability determination threshold When comparing, Furthermore, if this state persists throughout the observation and evaluation period, the cooperative region is determined. The lack of effective observations from participants in the space-air-ground collaboration has led to the region being marked as an unobservable collaboration region. The marking status of the unobservable collaboration region indicates that the current state of the region cannot be directly obtained from real-time observation information. After a collaborative region is marked as an unobservable collaborative region, the space-air-ground collaborative layer no longer directly relies on real-time observation results to update the state of the region. Instead, it combines the adjacent state information of the collaborative key region state and historical state records to infer and update the current state of the unobservable collaborative region, so as to avoid the state interruption of the collaborative key region state under unobservable conditions. To perform state inference updates, the space-air-ground coordination layer first acquires the state of key coordination regions that are spatially or in coordination relationships with the unobservable coordination region. The set of adjacent key coordination regions is denoted as... ,in Includes unobservable cooperative regions in terms of spatial location, cooperative tasks, or state relationships. For key collaborative regions with adjacency relationships, the space-air-ground collaborative layer extracts the corresponding state attribute values ​​from the states of adjacent key collaborative regions as spatial or collaborative association reference information. The space-air-ground coordination layer simultaneously acquires historical state records of the unobservable coordination region before it enters an unobservable state. These historical state records can be represented as a sequence of states formed within the most recent observable time period. ,in Represents the historical state sampling time and satisfies , Indicates the collaborative region at this moment. The state value, This indicates the length of the historical state record, which is used to characterize the state evolution trend of the unobservable cooperative region before it enters the unobservable state. The space-air-ground coordination layer infers and updates the current state of unobservable coordination areas based on the correlation between the states of adjacent key coordination areas and historical state records. This state inference can be achieved through a weighted fusion method. As one possible approach, it updates the current state of unobservable coordination areas at any given time. The inferred state is represented as: ; in Indicates collaborative region At any moment The inferred state, Indicates the state of adjacent collaborative key regions at time [time]. The weighted average, This represents the reference state value calculated based on historical state records. Describes the state fusion weight coefficients and satisfies The fusion weighting coefficient is used to balance the influence of adjacent regional state information and historical state information on state inference; The space-air-ground coordination layer writes the inferred unobservable coordination region state into the corresponding state attribute in the coordination key region state, so that the unobservable coordination region still has a state expression that can be used by subsequent coordination decision-making and task generation processes during the coordination operation, thereby ensuring the continuous operation of the space-air-ground coordination process in the presence of observation gaps.

[0034] S6. Based on the status of key collaborative areas, collaborative responsibility status, status of unobservable collaborative areas, and collaborative decision inputs, generate collaborative tasks for airborne UAVs and issue them for execution.

[0035] Furthermore, in S6, the collaborative tasks for generating airborne UAVs include: Based on the collaborative responsibility status, determine the collaborative role that the airborne UAV plays in the current collaborative phase; By combining the status of key collaborative areas and the status of unobservable collaborative areas, the target areas corresponding to collaborative roles are determined from the key collaborative areas. Based on the state characteristics of the target area, generate airborne UAV collaborative missions that match the collaborative roles.

[0036] Specifically, after obtaining the status of key collaborative areas, collaborative responsibility status, inferred status of unobservable collaborative areas, and collaborative decision input, the air-space-ground collaborative layer generates collaborative tasks for airborne UAVs based on the above information and distributes the generated collaborative tasks to the corresponding airborne UAVs for execution, enabling the airborne UAVs to participate in air-space-ground collaborative operation in the current collaborative phase according to the collaborative decision results. In the specific implementation process, the collaborative responsibility status is used to indicate the division of decision-making and execution tasks among the collaborative participants in the current collaborative phase. The air-space-ground collaborative layer determines the collaborative role undertaken by the airborne UAV in the current collaborative phase based on the collaborative responsibility status. The collaborative role is used to characterize the functional positioning of the airborne UAV in collaborative operation. The collaborative role may include, but is not limited to, observation compensation role, status verification role, area coverage role, or collaborative execution role. The determination of the collaborative role is used to constrain the target type and behavior mode generated by the airborne UAV in subsequent tasks. After determining the collaborative role of the airborne UAV, the air-space-ground collaboration layer combines the status of key collaborative areas and the status of unobservable collaborative areas to determine the target area corresponding to the collaborative role from the key collaborative areas. The target area is used to indicate the area range that the airborne UAV needs to focus on or perform tasks in the current collaborative phase. The determination of the target area is based on the description of the status characteristics of each area in the status of key collaborative areas, and the inferred status of unobservable collaborative areas is used as an important basis for the selection of target areas, so that the target area can cover the areas where the status of the current collaborative operation is insufficient or needs to be supplemented. The determination of the target area can be achieved by matching the collaborative role with the regional state characteristics. The regional state characteristics can include information such as the uncertainty level of the area, the observation availability status, the state change trend or the task correlation. The air-space-ground collaboration layer selects areas that meet the preset conditions from the key collaborative areas as the target area set according to the functional requirements of the collaborative role, and prioritizes the target areas when needed to support task generation in multi-objective scenarios. After obtaining the target area, the air-space-ground coordination layer generates an airborne UAV coordination task that matches the coordination role based on the state characteristics of the target area. The coordination task describes the specific actions or behavioral constraints that the airborne UAV needs to perform in the target area. The coordination task can include information such as task type, target area identifier, task execution parameters and task constraints. The generation of the coordination task uses the coordination decision input as a constraint to ensure that the task content is consistent with the coordination decision of the current coordination stage. To map regional states to specific task content, the space-air-ground coordination layer can construct task generation mapping relationships. As one possible approach, this involves mapping the target region... The corresponding collaborative task is represented as follows: ; in Indicates targeting area The generated airborne UAV collaborative mission, Indicates targeting area Collaborative decision-making input, Indicates the target area The corresponding collaborative responsibility status, Indicates the target area The current state of the region, where the target region belongs to an unobservable cooperative region. This represents the region state obtained through inference and updating. This represents a task generation mapping function, used to jointly map decision inputs, responsibility status, and region status into executable collaborative tasks; The task generation mapping function is used to comprehensively process different input elements. The task generation mapping function can include steps such as task type selection, task parameter configuration and task constraint generation. The specific implementation of the task generation mapping function can be configured according to the mission capabilities, control interface or task execution model of the airborne UAV, so that the generated collaborative task can be directly parsed and executed by the airborne UAV. After generating the collaborative mission for airborne UAVs, the air-space-ground collaboration layer distributes the collaborative mission to the corresponding airborne UAVs. The airborne UAVs execute corresponding flight control, perception operations, or collaborative behaviors based on the received collaborative mission, and continuously feed back the execution status to the air-space-ground collaboration layer during the mission execution process. This allows the status of key collaborative areas to be updated as the mission execution process progresses, forming a closed-loop process of air-space-ground collaborative operation.

[0037] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for UAV air-ground collaborative operation, characterized in that, Includes the following steps: S1. Construct the state of key collaborative areas in the space-air-ground collaborative layer, making them independent of any single data source; S2. Write the collaboration-related information from the participants in the air-space-ground collaboration into the corresponding status attributes of the collaborative key area status according to the preset attribute classification method, and handle the conflict of update requests from different sources for the same status attribute, and write the processing results into the collaborative key area status. S3. Based on the state of the key collaborative area, characterize the state uncertainty of the participants in the air-space-ground collaboration within the key collaborative area, and determine the responsibility relationship for collaborative decision-making in the air-space-ground collaboration process based on the state uncertainty, thereby generating the collaborative responsibility state. S4. Determine the collaborative decision-making entity based on the collaborative responsibility status, and generate collaborative decision input based on the status information with low uncertainty in the collaborative key area status; S5. Based on the available observation information of the participants in the air-space-ground collaboration, identify areas that cannot be effectively observed and mark them as unobservable collaborative areas. Combine the adjacent state information and historical state records of the key collaborative areas to infer and update the state of the unobservable collaborative areas. S6. Based on the status of key collaborative areas, collaborative responsibility status, status of unobservable collaborative areas, and collaborative decision inputs, generate collaborative tasks for airborne UAVs and issue them for execution.

2. The UAV air-ground collaborative method according to claim 1, characterized in that, In S1, the construction of the collaborative key region state includes: In the space-air-ground collaborative layer, identify key areas as objects of collaborative concern; Establish corresponding state description structures for key areas to characterize the state attributes of key areas in the air-space-ground coordination process; By decoupling the state description structure from the participants in the space-air-ground collaboration, the state of the key collaborative areas can exist independently of any single data source and be jointly updated by the participants in the space-air-ground collaboration.

3. The UAV air-ground collaborative method according to claim 1, characterized in that, In S2, the step of writing the corresponding state attributes of the collaborative key area state according to the preset attribute division method includes: The relevant information from the participants in the air-space-ground collaboration is classified by attribute type, and the relevant information is written into the mutually isolated state attribute set in the state of the key collaborative area according to the classification result; Different state attribute sets are used to characterize the state changes of key areas under different collaborative dimensions, and the collaborative information written into different state attribute sets does not participate in the same update competition in subsequent state update processes.

4. The UAV air-ground collaborative method according to claim 1, characterized in that, In S2, the conflict handling for update requests from different sources for the same state attribute includes: Within the same state update cycle, detect whether multiple update requests for the same state attribute exist simultaneously; When multiple update requests are detected, the priority of the multiple update requests is determined by combining the collaborative responsibility status; Based on the priority determination results, full writes are performed on update requests with higher priority, while restricted writes or delayed writes are performed on update requests with lower priority.

5. A method for UAV air-ground collaborative operation according to claim 1, characterized in that, In S3, the uncertainty in the state of the participants in the air-space-ground collaboration within the key collaborative area includes: Within a preset state evaluation time window, the update frequency of each state attribute is statistically analyzed to characterize the degree of uncertainty in the changes of state attributes. Calculate the magnitude of change of each state attribute within the state evaluation time window to characterize the degree of uncertainty of the state attribute changes; The correlation changes between different state attributes within the state evaluation time window are analyzed to determine the interaction relationships between state attributes, which are used to characterize the joint uncertainty between state attributes.

6. The UAV air-ground collaborative method according to claim 1, characterized in that, In S3, the determination of the collaborative decision-making relationship in the air-space-ground coordination process includes: Based on state uncertainty, calculate the information entropy of the participants in the air-space-ground collaboration within the key collaborative area; Compare the information entropy of different participants in the air-space-ground collaboration and select the participant with the lower information entropy as the decision-maker for the current collaboration stage. When the information entropy changes and the preset collaborative adjustment conditions are met, the collaborative decision-making relationship is adjusted and the collaborative responsibility status is updated.

7. The UAV air-ground collaborative method according to claim 1, characterized in that, In S4, generating collaborative decision input based on state information with low uncertainty in the collaborative key area state includes: Based on the collaborative responsibility status, select the status attributes corresponding to the current collaborative decision-making relationship from the status of key collaborative areas; The selected state attributes are reorganized to generate a decision input structure that meets the processing requirements of the collaborative decision generation subject; The decision input structure is provided as collaborative decision input to the collaborative decision generation entity.

8. The UAV air-ground collaborative method according to claim 1, characterized in that, In S5, determining and marking regions where effective observations cannot be formed as unobservable cooperative regions includes: During the collaborative operation, observation availability information for the collaborative area is obtained from each of the participants in the air-space-ground collaboration. Based on observation availability information, determine whether the collaborative area continues to lack effective observations from space-air-ground collaborative participants during the preset observation evaluation period. When a cooperative region is determined to have a persistent lack of effective observations, it is marked as an unobservable cooperative region.

9. A method for UAV air-ground collaborative operation according to claim 1, characterized in that, In S5, the step of inferring and updating the state of unobservable cooperative regions by combining the adjacent state information and historical state records of the cooperative key regions includes: Obtain the status of key collaborative regions that are spatially or in collaborative relationships with unobservable collaborative regions; Obtain historical state records of the unobservable cooperative region before it enters an unobservable state; Based on the correlation between the state of adjacent collaborative key areas and historical state records, the current state of unobservable collaborative areas is inferred and updated.

10. A method for UAV air-ground collaborative operation according to claim 1, characterized in that, In S6, the cooperative task of generating the airborne UAV includes: Based on the collaborative responsibility status, determine the collaborative role that the airborne UAV plays in the current collaborative phase; By combining the status of key collaborative areas and the status of unobservable collaborative areas, the target areas corresponding to collaborative roles are determined from the key collaborative areas. Based on the state characteristics of the target area, generate airborne UAV collaborative missions that match the collaborative roles.

Images