A method and system for working of IOODA architecture of swarm intelligence cooperation

Abstract

The application discloses an IOODA architecture working method and system of cluster intelligence cooperation, relates to the field of aircraft cluster control technology, and comprises the following steps: calculating the threat degree of an aircraft to a target according to the state information of the aircraft and basic information of the target; calculating the target allocation result of the aircraft based on the threat degree and an initial allocation vector of the target; calculating the task allocation result of the aircraft based on a task cost function matrix corresponding to a target group and a tentative allocation vector of the task; calculating the cooperative trajectory of the aircraft in the target group based on the basic information of the target corresponding to the target group, the number of aircraft in the target group and the state information of the aircraft; and performing a corresponding flight task based on the cooperative trajectory of the aircraft. The application solves the problems of information interaction loss, insufficient autonomy and dependence on pre-prepared programs in the current cluster cooperation process, and realizes efficient, reliable and self-organizing cooperative operation capability.

Description

Technical Field

[0001] This application relates to the field of aircraft swarm control technology, and in particular to a working method and system for an IOODA architecture for intelligent swarm collaboration. Background Technology

[0002] With the rapid development of unmanned systems technology, multi-aircraft swarm cooperative control has become a research hotspot in the aerospace field. In military applications, swarm cooperative operations can significantly improve mission execution efficiency, enhance penetration capabilities, and improve system robustness. A swarm system refers to a comprehensive equipment system consisting of a certain number of aircraft interconnected through a communication network, relying on a distributed coordination mechanism to achieve tasks such as cooperative reconnaissance and cooperative strike. Currently, research on swarm systems has accumulated considerable achievements, mainly focusing on individual technologies such as cooperative perception, cooperative decision-making, cooperative control, and cooperative guidance, as well as system composition design.

[0003] Existing technologies and real-world examples of drone swarm collaborative operations primarily exhibit novel models such as "order-based" operations. However, in these application scenarios, the swarm collaboration process still heavily relies on macro-level control by the "human-in-the-loop" (HIL), with ground stations centrally commanding drones to execute missions. Information exchange between drones is relatively limited, resulting in insufficient autonomy and adaptability in overall collaboration. Moreover, in existing publicly available technologies, drone swarm collaborative operations largely depend on pre-programmed collaboration schemes, meaning that most of the collaboration logic is already embedded in the system before mission execution. The response processes of each OODA loop are basically pre-set, lacking the ability to autonomously adjust according to dynamic environments and mission changes. Simultaneously, information exchange (the "I" loop) is mostly limited to one-way command transmission from the ground station to the drones. There is a lack of substantial, multi-dimensional information exchange and collaborative processing between nodes within the swarm, resulting in limited flexibility and robustness of the system as a whole when dealing with uncertainties and complex tasks. Summary of the Invention

[0004] The purpose of this application is to provide a working method and system for an IOODA architecture of intelligent cluster collaboration. Through substantial and multi-dimensional information exchange and collaborative processing among various aircraft nodes within the cluster, the cluster can achieve independent and autonomous collaborative task execution, breaking through the limitations of pre-programmed collaboration.

[0005] To achieve the above objectives, this application provides the following solution: In a first aspect, this application provides a working method for an IOODA architecture of intelligent cluster collaboration, characterized in that the cluster includes multiple aircraft; each aircraft is used to execute the working method of the IOODA architecture of intelligent cluster collaboration; the working method of the IOODA architecture of intelligent cluster collaboration includes the following steps.

[0006] The system acquires in real time the status information of all aircraft corresponding to the target area and the basic information of all targets within the target area; the status information includes position information and velocity information; the basic information includes status information and category information.

[0007] Based on the status information of its own aircraft and the basic information of all targets within the target area, the threat level of its own aircraft to each target is calculated.

[0008] Based on the threat level of the aircraft to each target and the initial target allocation vectors output by all aircraft corresponding to the target area, the target allocation results of the aircraft are calculated and output.

[0009] Based on the task cost function matrix corresponding to the target group and the provisional task allocation vector output by all aircraft in the target group, the task allocation result corresponding to its own aircraft is calculated and output; the target group is determined according to the target allocation result; all aircraft in the target group correspond to the same target.

[0010] Based on the basic information of the target corresponding to the task group and the number of aircraft in the task group, the mission terminal location information corresponding to its own aircraft is determined, and based on the mission terminal location information corresponding to its own aircraft and the status information of all aircraft in the task group, the virtual flight time of its own aircraft is determined; the task group is determined according to the task allocation result, and all aircraft in the task group correspond to the same task.

[0011] Based on the status information of its own aircraft within the task group, the location information of the corresponding task terminal of its own aircraft, and the virtual flight time of its own aircraft, the cooperative trajectory corresponding to its own aircraft within the task group is calculated, and the corresponding flight mission is executed based on the cooperative trajectory corresponding to its own aircraft.

[0012] Secondly, this application provides a system based on IOODA control, characterized in that the system based on IOODA control includes multiple aircraft; each of the aircraft is used to execute the steps of the above-described cluster intelligent collaborative IOODA architecture working method.

[0013] According to the specific embodiments provided in this application, this application has the following technical effects: This application addresses the urgent need for independent and autonomous collaborative task execution in cluster systems, aiming to overcome the limitations of pre-programmed collaboration and proposing a feasible IOODA architecture working method and system for intelligent cluster collaboration. It systematically presents a generalized logical flow for unmanned cluster collaborative task execution, clearly depicting the entire signal flow process from information perception and interaction processing to collaborative decision-making and action, achieving deep integration of information interaction with the OODA (Observation-Judgment-Decision-Action) cycle. Furthermore, for the more complex and dynamic task scenarios faced by cluster systems, this application further proposes intelligent management methods covering all stages of IOODA, including dynamic task allocation, autonomous collaborative decision-making, distributed information fusion, and real-time adaptive adjustment mechanisms. This solves problems such as lack of information interaction, insufficient autonomy, and reliance on pre-programmed procedures in current cluster collaboration processes, thereby supporting clusters to achieve efficient, reliable, and self-organizing collaborative operation capabilities with minimal human intervention. Attached Figure Description

[0014] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0015] Figure 1 A flowchart illustrating the working method of an IOODA architecture for clustered intelligent collaboration, provided in an embodiment of this application; Figure 2 This is a schematic diagram of the spatial geometric relationship of aircraft in a cluster provided in an embodiment of this application; Figure 3 This is a schematic diagram of the spatial geometric relationship of a formation configuration provided in one embodiment of this application; Figure 4 This application provides a schematic diagram of the IOODA brain-based distributed state machine logic operation principle as an embodiment of the present application. Figure 5 This is a hardware structure diagram of the IOODA brain provided in an embodiment of this application; Figure 6 This is a diagram illustrating the IOODA brain distributed state machine and its hardware structure according to an embodiment of this application. Figure 7 This is a schematic diagram of the composition structure of the IOODA Brain Logic Intelligent Operation Management System provided in an embodiment of this application. Detailed Implementation

[0016] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0017] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0018] In one exemplary embodiment, such as Figure 1 As shown, a method for operating an IOODA architecture for cluster intelligent collaboration is provided, wherein the cluster includes multiple aircraft. In this embodiment, each aircraft is used to execute the method for operating an IOODA architecture for cluster intelligent collaboration; the method includes steps 101 to 106.

[0019] Step 101: Real-time acquisition of status information of all aircraft corresponding to the target area and basic information of all targets within the target area; the status information includes position information and velocity information; the basic information includes status information and category information.

[0020] Step 102: Based on the status information of its own aircraft and the basic information of all targets in the target area, calculate the threat level of its own aircraft to each target.

[0021] Step 103: Based on the threat level of the aircraft to each target and the initial target allocation vectors output by all aircraft corresponding to the target area, calculate and output the target allocation results of the aircraft.

[0022] Step 104: Based on the task cost function matrix corresponding to the acquired target group and the provisional task allocation vector output by all aircraft in the target group, calculate and output the task allocation result corresponding to the aircraft itself; the target group is determined according to the target allocation result; all aircraft in the target group correspond to the same target.

[0023] Step 105: Based on the basic information of the target corresponding to the task group and the number of aircraft in the task group, determine the task terminal location information corresponding to the aircraft itself, and based on the task terminal location information corresponding to the aircraft itself and the status information of all aircraft in the task group, determine the virtual flight time of the aircraft itself; the task group is determined according to the task allocation result, and all aircraft in the task group correspond to the same task.

[0024] Step 106: Based on the status information of its own aircraft within the task group, the location information of the task terminal corresponding to its own aircraft, and the virtual flight time of its own aircraft, calculate the cooperative trajectory corresponding to its own aircraft within the task group, and execute the corresponding flight mission based on the cooperative trajectory corresponding to its own aircraft.

[0025] This application addresses the urgent need for independent and autonomous collaborative task execution in cluster systems. It aims to overcome the limitations of pre-programmed collaboration by implementing steps 101 to 106, proposing a feasible IOODA architecture working method and system for intelligent cluster collaboration. The application systematically presents a generalized logical flow for unmanned cluster collaborative task execution, clearly depicting the entire signal flow process from information perception and interaction processing to collaborative decision-making and action, achieving deep integration of the information interaction stage with the OODA (Observation-Judgment-Decision-Action) cycle. Furthermore, for the more complex and dynamic task scenarios faced by cluster systems, this application further proposes an intelligent management and control method covering all stages of IOODA, including dynamic task allocation, autonomous collaborative decision-making, distributed information fusion, and real-time adaptive adjustment mechanisms. This solves problems such as lack of information interaction, insufficient autonomy, and reliance on pre-programmed procedures in current cluster collaboration processes, thereby supporting clusters to achieve efficient, reliable, and self-organizing collaborative operation capabilities with minimal human intervention.

[0026] In one exemplary embodiment, a working method for a cluster-intelligent collaborative IOODA architecture is provided, including the following steps.

[0027] I. Grouping of Aircraft Clusters The cluster comprises multiple aircraft; and a communication and interaction architecture is established for all aircraft in the cluster, with each aircraft executing the IOODA architecture working method for intelligent collaboration within the cluster.

[0028] In an exemplary embodiment, when the cluster arrives at the mission area and enters the target cooperative identification phase, the cluster uses a segmented search approach to perform target cooperative search and detection. Specifically, the mission area is divided into multiple target areas, and the aircraft cluster is assigned to each target area. Each target area contains one aircraft group, totaling [number missing]. A group of aircraft, each group including One aircraft, , A single aircraft in the target area Search along different altitudes, and The target area Includes target quantity One. To ensure no area is missed in the search, a cross-searching and repeated search method is used. The aircraft group is in the target area. Conduct the first fixed route search ( During the search, the aircraft You will encounter One goal ( ; ), the first aircraft You will encounter One goal ( ; ).

[0029] In one exemplary embodiment, the task region is not segmented, and the entire task region is a single target region.

[0030] The following steps all take any target area (aircraft group) as an example.

[0031] II. Collaborative Target Identification The following steps are performed for each aircraft.

[0032] The method for obtaining the category information specifically includes the following steps.

[0033] Step 201: Obtain the set of observation images of all targets detected within the target area, and perform target recognition based on the set of observation images to obtain and output the preliminary recognition results of all targets identified by the aircraft; the preliminary recognition results include preliminary confidence, preliminary category information and preliminary feature descriptors.

[0034] Target recognition based on a set of observed images includes the following steps.

[0035] Step 2011: The aircraft performs target detection within the target area, acquires a set of observation images of all targets detected by the aircraft within the target area, and performs target recognition based on the set of observation images of all targets detected by the aircraft and a target detection network (e.g., YOLOv8 target detection network) to obtain a set of attributes of all targets recognized by the aircraft.

[0036] The attribute set is represented as follows: ; in; Indicates aircraft Identified target The set of attributes; Indicates aircraft Identified target The detection frame; Indicates aircraft Identified target The initial confidence level; Indicates aircraft Identified target Preliminary category information; Indicates aircraft The identified set of targets; This represents an object detection network; Indicates aircraft The collection of identified observation images; .

[0037] Step 2012: The detection bounding boxes of the attribute set of each target identified by the aircraft are cropped and preprocessed to obtain the processed local appearance image of each target identified by the aircraft.

[0038] The cutting process includes: ; in, Indicates the aircraft Identified target A cropped partial appearance image; This indicates an image cropping operation, which extracts the target region from the original image based on the target's detection bounding box.

[0039] The preprocessing process includes processing the local appearance image after the target is cropped using one or more of the following methods: size normalization, contrast enhancement, and noise reduction.

[0040] Step 2013 involves using a trained feature extraction network to extract features from the processed local appearance image of each target identified by the aircraft, obtaining a preliminary feature descriptor for each target. This yields a preliminary identification result for each target, including preliminary confidence level, preliminary category information, and a preliminary feature descriptor. The preliminary feature descriptor for each target identified by the aircraft serves as the target's feature information for information exchange and collaborative identification among the aircraft during target recognition.

[0041] The trained feature extraction network is as follows: ; in, Indicates the aircraft Identified target Feature descriptors; This represents the trained feature extraction network.

[0042] The trained feature extraction network is obtained by using multi-view images and contrastive learning.

[0043] The trained feature extraction network is obtained by using multi-view images and contrastive learning, specifically including the following steps.

[0044] To learn viewpoint-invariant feature representations, ensuring that the same target has similar features across different viewpoints, while different targets exhibit distinct features, a positive and negative sample set needs to be constructed when training the feature extraction network (which could be a CNN, ViT, etc.). The positive sample set consists of observation images of the same target from different aircraft. The negative sample set consists of randomly sampled observation images of any target observed by other aircraft. Each target region corresponds to an aircraft group, and the other aircraft are all aircraft in the group except for the target itself.

[0045] The feature extraction network is trained using an improved InfoNCE loss function, which is as follows: ; in, Indicates aircraft and aircraft Detection consistency loss function; This represents an exponential function with base e. Represent a similarity function; This represents an adjustable constant. Indicates aircraft Identified target Preliminary feature descriptor; Indicates aircraft Identified target Preliminary feature descriptor; and This refers to different aircraft within the same target area; each target area corresponds to a group of aircraft. This refers to all aircraft in the group except for its own aircraft.

[0046] During the training of the feature extraction network, the improved InfoNCE loss function can be weighted using information such as the initial confidence level of the target and the initial category information of the target, resulting in a weighted contrastive loss function.

[0047] The weight matching formula is as follows: ; in, Indicates aircraft and aircraft For the identified target The matching weights; σ represents the activation function, and MLP represents the multilayer perceptron; Indicates aircraft Identified target The initial confidence level; Indicates aircraft Identified target The initial confidence level; Indicates aircraft Identified target Preliminary category information; Indicates aircraft Identified target Preliminary category information.

[0048] The weighted contrast loss function is: ; in, This represents the weighted contrastive loss function for collaborative detection; The feature extraction network is trained using this adaptive weighted contrastive loss function, resulting in a well-trained feature extraction network.

[0049] Step 202: Based on the preliminary target identification results output by all aircraft corresponding to the acquired target area, calculate the similarity between the preliminary identification results of the first target and the preliminary identification results of the second target. When the similarity is greater than a first threshold, determine that the first target is an independent target; when the similarity is less than or equal to the first threshold, determine that the second target is a similar target of the first target, thus obtaining a set of similar targets of the first target. The first target is any target identified by the aircraft itself, and the second target is any target identified by other aircraft; the first threshold is a small numerical value.

[0050] The system acquires preliminary target identification results from all aircraft corresponding to the target region, including preliminary identification results of targets identified by the aircraft itself and preliminary identification results of targets output by other aircraft. The target region corresponds to an aircraft group, and the other aircraft are all aircraft in the group except the aircraft itself. The similar target set includes the similarity between the preliminary identification results of the first target and the preliminary identification results of all second targets, as well as the preliminary identification results of all targets whose similarity to the preliminary identification result of the first target is greater than a first threshold.

[0051] The formula for calculating similarity is: ; in, Indicates aircraft Detected With aircraft Detected The similarity.

[0052] Step 203: Based on the set of similar targets of the first target and the preliminary identification results of the first target, the preliminary feature descriptor is corrected to obtain the corrected feature descriptor, which specifically includes the following steps.

[0053] The modified feature descriptor of the target is obtained by weighting the initial confidence and similarity of all targets in the set of similar targets of the first target.

[0054] The formula for calculating the modified feature descriptor is as follows: ; in, Indicates aircraft Identified target Modified feature descriptors; Indicates aircraft Identified target The corresponding set of similar targets; Indicates aircraft Identified target Aircraft in the corresponding set of similar targets Identified target Similarity; Indicates aircraft Identified target The initial confidence level; Indicates aircraft Identified target Preliminary feature descriptor; Indicate target Aircraft in the corresponding set of similar targets Identified target The initial confidence level; Indicate target Aircraft in the corresponding set of similar targets Identified target Preliminary feature descriptor.

[0055] This application uses a weighted calculation based on confidence and similarity to aggregate feature information of similar targets in other aircraft for any target in any aircraft, thereby obtaining a modified feature descriptor for the target.

[0056] Step 204: Based on the revised feature descriptor, determine the category information of the corresponding target. This specifically includes the following steps: The modified feature descriptors are then used for final target identification to determine the target's category information and confidence level, and the target identification results are updated. In this way, the aircraft completes cooperative target identification and obtains the final target identification result.

[0057] The final target identification results of any aircraft may contain duplicate targets with those of other aircraft. Therefore, duplicate targets detected by all aircraft are removed.

[0058] Randomly select one aircraft from all aircraft corresponding to the target area. Merge the final identification results of all aircraft identifying targets, eliminating duplicate targets (for example, a voting method can be used for duplicate elimination; that is, for the same candidate target, if more than half of the aircraft identify it as the same target, it is judged as a duplicate and merged). By eliminating duplicate targets, the final identification results of all targets are obtained and output. The final identification result includes the target category information.

[0059] It also includes step 205, in which the aircraft cluster conducts a second fixed-route search, assigns a number to the final target, and outputs the observation and detection results. This step further confirms the first flight.

[0060] (III) Target Collaboration and Relationship Perform the following steps for each aircraft: The acquisition of basic information on all targets within the target area includes the following steps.

[0061] Step 301: Obtain the position information of the aircraft itself and the observation information of the target identified by the aircraft itself, calculate the target position information, and obtain the first position information of the target identified by the aircraft itself; specifically, it includes the following steps.

[0062] Step 3011: The aircraft acquires its own status information, which includes position information and speed information.

[0063] The aircraft's position information is represented as three-dimensional position information. ,in Indicates aircraft exist The coordinates of the axes; where Indicates aircraft exist The coordinates of the axes; where Indicates aircraft exist The coordinates of the axes. The velocity information of the aircraft is represented as three-dimensional velocity information. ,in Indicates aircraft exist Velocity in the axial direction; where Indicates aircraft exist Velocity in the axial direction; where Indicates aircraft exist Velocity in the axial direction.

[0064] Step 3012: Obtain observation information of targets identified by the aircraft itself.

[0065] The observation information includes the relative distance, pitch angle, and azimuth angle between the aircraft and the identified target.

[0066] Observational information can be represented as ;in, ; Indicates aircraft The set of all identified targets; Indicates aircraft With the identified target The relative distance; Indicates aircraft With the identified target The azimuth angle; Indicates aircraft With the identified target The pitch angle.

[0067] Observational information can be used as follows Figure 2 express, For a common coordinate system, For a local coordinate system, the origin is... For any aircraft The target is any target .definition coordinate system and The rotation relationship of the coordinate system is as follows: . Indicates aircraft With the identified target The relative distance; Indicates aircraft With the identified target The azimuth angle; Indicates aircraft With the identified target The pitch angle.

[0068] Unless otherwise stated, all location information in this application refers to coordinates in a common coordinate system.

[0069] Step 3013: The target's first location information is determined.

[0070] Based on the acquired position information of its own aircraft, the position information is represented as: That is, the above-mentioned acquisition .

[0071] Based on the observation information of the target identified by the aircraft itself, the coordinates of the target in the local coordinate system are obtained, which can be represented as: ; in, Indicates aircraft Identified target In the local coordinate system The coordinates of the axis; Indicates aircraft Identified target In the local coordinate system The coordinates of the axis; Indicates aircraft Identified target In the local coordinate system The coordinates of the axis.

[0072] Based on the coordinates of the target identified by the aircraft in the local coordinate system and the rotation relationship between the common coordinate system and the local coordinate system, the position information of the target identified by the aircraft is calculated. The first position information of the target identified by the aircraft is the coordinate information in the common coordinate system.

[0073] The formula for calculating target location information is: ; in, Indicates aircraft The rotation relationship between the local coordinate system and the common coordinate system.

[0074] Repeat the above steps to iterate and calculate the observation information of all targets identified by the aircraft; obtain the first position information of all targets identified by the aircraft and output it.

[0075] Step 302: Randomly select one aircraft from all aircraft corresponding to the target area, obtain the first position information of the identified target output by all other aircraft, fuse the first position information of the target identified by all aircraft, extract the first position information of any target in the fused information and calculate the average value to obtain the rough position information of all targets in the target area.

[0076] Iterate through all targets in the fused information to obtain and output the approximate location information of all targets within the target area.

[0077] Step 303: Based on the position and velocity information of the aircraft itself, as well as the rough position and observation information of the target identified by the aircraft itself, perform local tracking operation. For example, use the Kalman filter tracking algorithm to perform local tracking operation, obtain the local tracking result of the target identified by the aircraft itself, and output it; the local tracking result is based on the tracking information of the aircraft itself on the identified target.

[0078] Step 304: Obtain the local tracking results output by all aircraft within the target area, perform filtering and cooperative tracking iterative operation, obtain the state information of the target identified by the aircraft itself, and output it; the state information includes position information and velocity information.

[0079] (iv) Threat Level Assessment Perform the following steps for each aircraft: Based on the status information of its own aircraft and the basic information of all targets in the target area, the threat level of its own aircraft to each target is calculated, which includes the following steps.

[0080] Step 401: Based on the status information of its own aircraft and the basic information of all targets in the target area, determine the threat element index of its own aircraft to each target; the threat element index includes type threat index, speed threat index, distance threat index and angle threat index.

[0081] The target category information in the basic information can be categories such as tanks, drones, and loitering munitions.

[0082] Step 4011, the step of determining the type threat index includes: Different types of targets pose different threats and will cause varying degrees of threat to aircraft. Experts assign type threat index values ​​based on the category information of different targets.

[0083] Based on the expert's assigned type threat index and the category information of all targets within the target area, the type threat index of the aircraft against each target is obtained. Indicates aircraft For the target The type of threat index.

[0084] Step 4012, the speed threat index determination step includes: To describe the threat posed by the target's speed during flight as comprehensively as possible, it is assumed that the target always desires to pose the greatest threat to the aircraft at an optimal speed, while the aircraft always desires to minimize the threat from the target at an appropriate speed.

[0085] Based on the speed information in the state information of the aircraft itself, the absolute speed of the aircraft is obtained; based on the absolute speed of the aircraft and the speed threat index formula, the speed threat index of the aircraft to each target is calculated.

[0086] The speed threat index is calculated using the following formula: ; in, Description of aircraft For the target Speed ​​threat index; Indicates aircraft The magnitude of the absolute velocity; Indicate target The absolute speed.

[0087] Step 4013, the distance threat index determination step includes: Typically, the threat posed by a target is a decreasing function of the relative distance to the aircraft.

[0088] Based on the position information in the status information of the aircraft itself, the position information in the basic information of all targets in the target area, and the distance threat index calculation formula, the distance threat index of the aircraft to each target is obtained.

[0089] The formula for calculating the distance threat index is: ; in, Description of aircraft For the target Distance threat index; Indicate target With aircraft The relative distance; Indicate target Location information; Indicate target exist The coordinates of the axis, Indicate target exist The coordinates of the axis, Indicate target exist The coordinates of the axis.

[0090] Step 4014, the step of determining the angle threat index includes: Consider the threat posed by the target from different angles. Assume the aircraft... With the goal The magnitude of the three-dimensional relative angle formed by the velocity directions is The smaller the relative angle, the greater the threat; therefore, it can be set to a linear inverse relationship. The three-dimensional relative angle is determined by the aircraft. Speed ​​information and target The speed information is determined.

[0091] Based on the speed information in the state information of the aircraft itself and the speed information in the basic information of all targets in the target area, the three-dimensional relative angle between the aircraft and all targets in the target area is determined. Based on the three-dimensional relative angle between the aircraft and all targets in the target area and the angle threat index calculation formula, the distance threat index of the aircraft to each target is obtained.

[0092] The formula for calculating the angle threat index is: ; in Indicates aircraft For the target Threat index from the angle; Indicates aircraft With the goal Three-dimensional relative angle; This represents a constant proportionality coefficient, the value of which is guaranteed to be... The result of the calculation is between 0 and 1.

[0093] Step 402: Perform a weighted summation of the threat element indices to obtain the relative threat likelihood probability of the aircraft to each target.

[0094] The weighted summation formula for relative threat likelihood probabilities is: ; in: Indicates aircraft For the target The relative threat likelihood probability; Indicates aircraft For the target The relative initial threat probability can be set to a uniform distribution (i.e., the relative initial threat probability of all targets is equal) or set according to expert experience; Indicates the first The weights of the various threat factor indices are determined by expert experience. .

[0095] This application primarily determines the threat level by calculating the posterior probability that a target poses a threat to an aircraft. This process is based on a weighted average of Bayes' theorem and threat factor indices.

[0096] Step 403: Randomly determine the initial relative threat probability of the aircraft to each target. Based on the relative threat likelihood probability and the initial relative threat probability, obtain the initial threat level of the aircraft to each target using Bayes' theorem.

[0097] Bayes' theorem is: ; in, Indicates aircraft For the target The initial threat level; Indicates aircraft For the target The initial threat level; Indicates aircraft For the target Relative initial threat probability; All target sets; It is a summation index used to iterate through all targets to ensure the sum of the posterior probabilities (for aircraft). () is standardized.

[0098] Step 404: Using the flight time, flight status of the aircraft, and flight status of the target as inputs, determine the correction factor using the threat correction factor model, and use the correction factor to correct the initial threat level of each target, thus obtaining the threat level of the aircraft to each target. The flight status includes three-dimensional position and three-dimensional velocity.

[0099] The threat level correction factor model is constructed as follows: Construct sample data, utilize expert experience, and for any given aircraft, at the [missing information]... In each sample, an aircraft was set up. For the target The correction factor at time 0 is 1, expressed as This means the threat level at this moment is the initial threat level. Through coordinated adversarial behavior formed based on the autonomous evolution of the target group, over time... ( After selecting several time points, experts will then adjust the correction factor. The threat weight time series was obtained by reassigning values ​​multiple times through experts. Sample. For this sample, the time can be obtained through curve fitting or LSTM-Transformer hybrid model fitting. A related threat level correction factor model makes , making , A large amount of sample data can be obtained through the threat level correction factor model to train the LSTM-Transformer hybrid model.

[0100] The threat level correction factor model can be used to obtain the self-vehicle threat level correction factor for any flight time.

[0101] Since the target nodes are not independent, there may be cluster task division and cluster collaborative confrontation behaviors, which will evolve over time. Therefore, the threat level of the target needs to be modified based on the initial threat level. The LSTM-Transformer hybrid model is used to perform hierarchical modification of the target threat level through a deep neural network architecture to obtain the final threat level, forming an intelligent intent recognition and evaluation system to judge the target's intent and understand the confrontation situation.

[0102] V. Target Coordination and Allocation (1) Conditions and constraints In the process of target allocation, it is necessary not only to consider maximizing the index function of the coordinated allocation of the following targets, but also to satisfy the realistic conditions, which are specifically manifested in the following two constraints: Constraint 1: A single aircraft may perform a mission against a single target at most.

[0103] Constraint 2: Each target must be assigned at least 2 aircraft to perform the mission.

[0104] (2) Target Coordination Allocation Perform the following steps for each aircraft: Based on the threat level of the aircraft to each target and the initial target allocation vectors output by all aircraft corresponding to the target area, the target allocation result of the aircraft is calculated, which includes the following steps.

[0105] Step 501: Randomly determine and output the initial target allocation vector of the aircraft.

[0106] The initial target assignment vector for the aircraft can be represented as: ; in Indicates aircraft Initial target assignment vectors for all targets; It is a binary variable. Indicates aircraft For the goal Serve; Indicates aircraft Not for the goal Serve, This indicates all targets within the target area.

[0107] Step 502: Obtain the initial target allocation vector for other aircraft within the target area.

[0108] Step 503 involves summing the initial target allocation vectors for the same target to obtain the target conflict perception of each target for the aircraft. This target conflict perception indicates how many aircraft are simultaneously competing for the same target; the larger this value, the more aircraft are competing for the same target.

[0109] The formula for calculating target conflict perception is: ; in, Indicates aircraft For the target Target conflict perception; This represents the collection of aircraft within the target area.

[0110] Step 504: Based on the aircraft's perception of target conflict for each target and the threat level of the aircraft to each target, calculate the aircraft's comprehensive target benefit for each target; comprehensive benefit refers to the quantitative assessment of the "net value" that the aircraft can obtain by serving a certain target after considering its own inherent conditions and external competitive pressure.

[0111] The formula for calculating the target comprehensive return is: ; Indicates aircraft For the target The overall target return Indicates aircraft For the target The level of threat; This represents a constant coefficient greater than zero, used to adjust the consensus conflict penalty term. Weight in the calculation of target comprehensive return; Indicates aircraft For the target Target conflict perception.

[0112] The formula for calculating the overall target return clearly shows: basic cost The lower the inherent threat level (i.e., the lower the cost), the greater the benefit. The higher the altitude, the more aircraft competing for the same objective. The larger the profit, the greater the benefit. The lower the punishment.

[0113] Step 505: Select the target corresponding to the maximum comprehensive benefit as the provisional target of the aircraft, calculate the provisional target allocation vector of the aircraft and output it.

[0114] Assuming the goal As an aircraft The provisional target is the aircraft. Target initial allocation vector update: And for all ,have The target assignment vector for the aircraft itself.

[0115] Step 506: Obtain the provisional target allocation vectors of other aircraft corresponding to the target area. Based on the provisional target allocation vectors of all aircraft corresponding to the target area, determine whether the provisional target allocation vectors of other aircraft are inconsistent with the provisional target allocation vector of the aircraft itself. If the provisional target allocation vectors of other aircraft are inconsistent with the provisional target allocation vector of the aircraft itself, it is determined that there is a conflict.

[0116] If so, the provisional target allocation vector of the aircraft itself is used as the initial target allocation vector of the aircraft itself, and the process returns to step 502.

[0117] If not, the provisional target of the aircraft is taken as the final target of the aircraft, and the target allocation result of the aircraft is obtained and output.

[0118] Step 507: Randomly select one aircraft from all aircraft corresponding to the target area, merge the target allocation results of all aircraft, obtain the target allocation results of all aircraft, and output them to all aircraft.

[0119] Through this iterative process of calculating local benefits and interacting with cluster information, the spacecraft group will eventually converge to a state where, for any target... At least one aircraft Make Furthermore, there is no conflict in the service objectives of all aircraft.

[0120] When the target assignment vectors of all aircraft remain unchanged, the system reaches a consensus on conflict-free task assignment, stops iterative calculation, and uses the result of the last iteration as the target assignment result.

[0121] VI. Task Coordination and Allocation Based on the task cost function matrix corresponding to the target group and the provisional task allocation vector output by all aircraft in the target group, the task allocation result corresponding to its own aircraft is calculated and output; the target group is determined according to the target allocation result; the target corresponding to all aircraft in the target group is the same target, which specifically includes the following steps.

[0122] Step 601: Establish the task cost function matrix corresponding to the task based on the obtained task type.

[0123] Step 602: Randomly determine the initial mission assignment vector for the aircraft itself.

[0124] The initial task allocation vector can be represented as: ; in Indicates aircraft The initial task assignment vector for all tasks corresponding to a given objective. It is a binary variable. Show aircraft Execute the task ; Show aircraft Do not perform task .

[0125] Step 603: Obtain the initial mission assignment vectors for other aircraft within the target group.

[0126] Step 604: Sum the results of the initial task allocation vectors for the same task to obtain the task conflict perception of the aircraft for each task.

[0127] Step 605: Based on the aircraft's perception of mission conflicts for each mission and the mission cost function matrix, calculate the aircraft's overall mission benefit for each mission.

[0128] Step 606: Select the task corresponding to the maximum overall task benefit as the provisional task of the aircraft, and calculate and output the provisional task allocation vector of the aircraft.

[0129] Step 607: Obtain the provisional mission allocation vectors of other aircraft corresponding to the target group. Based on the provisional mission allocation vectors of all aircraft corresponding to the target group, determine whether the provisional mission allocation vectors of other aircraft conflict with the provisional mission allocation vectors of the target group itself.

[0130] If so, the provisional mission allocation vector of the aircraft itself is used as the initial mission allocation vector of the aircraft itself, and the process returns to step 603.

[0131] If not, the provisional mission of the aircraft itself will be taken as the final mission of the aircraft itself, and the mission allocation result of the aircraft itself will be obtained.

[0132] Step 608: Randomly select one aircraft from all aircraft in the target group, merge the task allocation results of all aircraft in the target group, obtain the task allocation results of all aircraft, and output them to all aircraft.

[0133] VII. Collaborative Trajectory Planning Based on the basic information of the target corresponding to the mission group and the number of aircraft within the mission group, determine the mission terminal location information corresponding to your own aircraft, specifically including: Step 701: After the task allocation is completed and the task allocation result corresponding to the aircraft is obtained, record the time when the task allocation is completed as the first moment. .

[0134] Step 702: Obtain the first-moment status information of the target corresponding to the task group and the first-moment status information of its own aircraft. The first-moment basic information includes the first-moment position information and the first-moment velocity information of the target. Step 703: Based on the acquired first-moment state information of the target and the formation configuration information of all aircraft in the mission group, determine the mission terminal position information.

[0135] The formula for determining the location of the task terminal is: ; in, Indicates aircraft The terminal location for executing the task The coordinates of the axis; Indicates aircraft The terminal location for executing the task The coordinates of the axis; Indicates aircraft The terminal location for executing the task The coordinates of the axis; This refers to any aircraft within the mission group; The first moment after the task assignment is completed; express Momentary Goal exist The coordinates of the axis; express Momentary Goal exist The coordinates of the axis; express Momentary Goal exist The coordinates of the axes; where Indicates aircraft Formation configuration information; Indicates aircraft relative elevation angles to the formation at the mission configuration center Indicates aircraft Yaw angle relative to the mission configuration center; Indicates aircraft Relative distance of the formation from the mission configuration center.

[0136] When the number of aircraft in the mission group is equal to 1, the position information of the target corresponding to the mission group is used as the mission terminal position information; at this time, no formation flight is performed, and the relative elevation angle between the aircraft and the mission configuration center, the relative yaw angle between the aircraft and the mission configuration center, and the relative distance between the aircraft and the mission configuration center are all 0.

[0137] When the number of aircraft in a mission group is greater than 1, and the aircraft in the mission group fly in formation throughout the entire process, the position information of the target corresponding to the mission group is used as the mission terminal position information; at this time, the relative elevation angle between the aircraft and the mission configuration center, the relative yaw angle between the aircraft and the mission configuration center, and the relative distance between the aircraft and the mission configuration center are all 0.

[0138] When there is more than one aircraft in the mission group and there is a target, the aircraft in the mission group perform formation configuration operations to determine the formation configuration information of each aircraft in the mission group. Based on the formation configuration information of each aircraft and the status of the target corresponding to the mission group, the mission terminal position information of each aircraft in the mission group is obtained.

[0139] Step 7031: Determine the formation configuration information for each aircraft within the mission group, specifically including: Randomly select one aircraft from the mission team to perform the following operations: Step 7031-1: Based on the acquired task type elements, obtain the overall fitness of the formation corresponding to the task. Specifically, this includes: Based on the type of mission elements (such as cooperative reconnaissance mission, elements include detection width, detection depth, inter-machine non-interference, maximum communication capability, etc.), determine the formation parameters corresponding to the mission and the fitness of each formation parameter. Then, perform a weighted summation of the fitness of all formation parameters to obtain the total fitness of the formation corresponding to the mission.

[0140] ; in, Indicates task The corresponding overall fitness of the formation; Indicates task Corresponding formation parameters The weight can be dynamically adjusted according to task requirements; Indicates task Corresponding formation parameters ; Indicates task The corresponding set of formation parameters.

[0141] Step 7031-2: Based on the obtained status information of all aircraft in the task group, calculate the position center of each aircraft in the task group, and calculate the mission configuration center based on the position center of each aircraft in the task group.

[0142] Step 7032-3: Randomly initialize the formation parameters of each aircraft in the mission group to obtain the initial formation parameters of each aircraft in the mission group.

[0143] The initial formation parameters include , , and , , ; in, , , This represents the relative distance between any two aircraft in three directions at time t=0. , , This represents the relative distance between any two aircraft in three directions at time t=1 second.

[0144] Step 7031-4: Based on the mission configuration center, the initial formation parameters of each aircraft in the mission group, and the gradient descent aircraft formation optimization algorithm, obtain the expected relative elevation angle, expected relative yaw angle, and expected relative distance between each aircraft in the mission group and the mission configuration center.

[0145] The gradient descent algorithm for optimizing aircraft formation is as follows: ; in, Indicates the first Its Jacobian matrix during iteration; Indicates the number of iterations ; Indicates the first Iterative flight Relative distance from the center of the mission configuration Indicates the first Iterative flight The relative elevation angle with respect to the center of the mission configuration Indicates the first Iterative flight The relative yaw angle with respect to the mission configuration center; Indicates the first The aircraft output after iteration Relative distance from the center of the mission configuration Indicates the first The aircraft output after iteration The relative elevation angle with respect to the center of the mission configuration Indicates the first The aircraft output after iteration The relative yaw angle with respect to the mission configuration center; This represents the overall fitness of the formation corresponding to the task.

[0146] When the difference between the output results of two consecutive iterations is less than the second threshold, that is... The iteration process is stopped, and the output of the last iteration is used as the expected relative elevation angle, expected relative yaw angle, and expected relative distance between the spacecraft and the mission configuration center. This represents the second threshold, a small parameter related to the task.

[0147] Step 7031-5: Based on the expected relative elevation angle, expected relative yaw angle, and expected relative distance between each aircraft in the mission group and the mission configuration center, as well as the formation configuration calculation formula, calculate and output the formation configuration information of each aircraft in the mission group.

[0148] The formula for calculating formation configuration is: ; in Indicates aircraft Formation configuration information, Indicates aircraft exist x Directional formation components, Indicates aircraft exist y Directional formation components; Indicates aircraft exist z Directional formation components.

[0149] To better understand the meaning of the parameters, such as Figure 3 As shown, suppose any aircraft in a mission group The desired location. Connecting the mission configuration center and the spacecraft. The desired position is determined by the length of the connecting line. Indicates aircraft The relative distance to the center of the mission configuration is expressed by the projection angle of the line in the common coordinate system. Indicates aircraft The relative elevation angle with respect to the center of the mission configuration is expressed by the projection angle of the line connecting them onto the common coordinate system. Indicates aircraft The relative yaw angle with respect to the mission configuration center, using the aircraft The relative distance from the mission configuration center, relative elevation angle, and relative yaw angle describe the aircraft in the mission configuration. The expected relative position relative to the mission centerline, which refers to the planned center reference trajectory.

[0150] Based on the location information of the mission terminal corresponding to its own aircraft and the first-moment status information of its own aircraft, the virtual flight time of its own aircraft is determined, which specifically includes the following steps: Step 801: Based on the mission terminal position information corresponding to the aircraft and the first moment state information of the aircraft, the flight time of the aircraft from the initial position to the mission terminal position is calculated and output using the three-dimensional proportional guidance method. The initial position is the position information of the aircraft at the first moment.

[0151] Step 802: Obtain the flight time from the initial position to the mission terminal position output by other aircraft in the mission group, and select the longest flight time as the virtual flight time of your own aircraft.

[0152] Randomly select one aircraft from all aircraft in the task group as the cooperative trajectory planning calculation node, and perform the following steps.

[0153] Step 901: Based on the status information of its own aircraft within the task group, the location information of its corresponding mission terminal, and its virtual flight time, calculate the cooperative trajectory of its own aircraft within the task group. This specifically includes the following steps: The cooperative trajectory includes a position cooperative trajectory and a velocity cooperative trajectory; the position cooperative trajectory includes the desired position command at each moment; the velocity cooperative trajectory includes the desired velocity command at each moment.

[0154] Step 9011: Based on the status information of the aircraft within the task group, the location information of the corresponding task terminal of the aircraft, and the virtual flight time of the aircraft, obtain the position coordination trajectory and the preliminary velocity coordination trajectory.

[0155] The location-coordinated trajectory includes shaft and Axis cooperative trajectory, and Axis-coordinated trajectory.

[0156] shaft and The axis cooperative trajectory is: ; in, for Indicates aircraft exist Position coordinates along the axis Indicates aircraft exist Position coordinates along the axis For polynomial coefficients, It is a sequence number, ranging from 0 to 6.

[0157] Specifically, polynomial coefficients There are 7 independent variables in the horizontal plane. shaft and Initial position constraints in two directions along the axis, and one initial velocity direction constraint. shaft and There are seven constraint variables: two terminal position constraints in two directions, one terminal velocity direction constraint, and one flight time constraint. These are obtained through parameter optimization methods (such as gradient descent). .based on Obtain the aircraft shaft and Axis trajectory function.

[0158] In the OXZ plane of the common coordinate system, design a uniform motion trajectory, then The axis cooperative trajectory is: ; in, Indicates aircraft exist Position coordinates along the axis; Indicates that the aircraft is in t g Moment z The position of direction; Indicates aircraft Virtual flight time; This refers to the current moment.

[0159] The initial velocity coordination trajectory is as follows: ; ; ; in, Indicates aircraft exist Desired velocity in the axial direction; Indicates aircraft exist Desired velocity in the axial direction; Indicates aircraft exist Desired velocity in the axial direction, Indicates the desired flight speed; / indicates an aircraft Slope of the trajectory.

[0160] Step 9012: Based on the virtual flight time of its own aircraft, the preliminary speed coordination trajectory is corrected to obtain the speed coordination trajectory.

[0161] The calculation formula for correcting the initial velocity-coordinated trajectory is as follows: ; The corrected velocity calculation formula is as follows: ; ; ; in, This represents the function for correcting flight speed control commands; The number of corrections (e.g., 20 times). It is a positive parameter. Indicates aircraft exist Desired speed for axial correction; Indicates aircraft exist Desired velocity correction in the axial direction; Indicates aircraft exist Expected speed correction in the axial direction.

[0162] Step 9013: Based on the position-coordinated trajectory and velocity-coordinated trajectory, determine and output the desired position and velocity commands for each aircraft within the mission group at each moment. This specifically includes the following steps: When the number of aircraft in the mission group is equal to 1, the expected position time function and the expected velocity time function are obtained by using the expected position and expected velocity calculation formulas. Based on the expected position time function and the expected velocity time function, the expected position command and expected velocity command of the aircraft at each moment are obtained.

[0163] The formulas for calculating the desired position and desired velocity are as follows: ; in, Indicates aircraft The expected position-time function; Indicates aircraft The expected velocity-time function; This refers to an aircraft performing a mission.

[0164] When the number of aircraft in the mission group is greater than 1, and the mission group needs to perform formation flight throughout the mission, an aircraft is randomly selected. The expected position and expected velocity of the mission configuration center of the mission group are obtained through the expected position and expected velocity calculation formulas. Based on the expected position time function and expected velocity time function, the expected position command and expected velocity command of the aircraft at each moment are obtained. The mission configuration center is the planned reference trajectory.

[0165] The formulas for calculating the desired position and desired velocity are as follows: ; in, The desired location time function representing the task configuration center of the task group; A time function representing the velocity information of the mission configuration center of the task group; This refers to an aircraft that executes the formulas for calculating the desired position and desired velocity.

[0166] When the number of aircraft in a mission group is greater than one, and the mission group needs to form a formation when a target is present, all aircraft in the mission group obtain their desired positions and velocities using the desired position and velocity calculation formulas. A randomly selected aircraft is chosen as the planning calculation node. After receiving the above instructions, it forwards the information to all aircraft in the mission group through the interaction module. Based on the desired position time function and the desired velocity time function, the desired position and velocity instructions for each aircraft at each moment are obtained.

[0167] The formulas for calculating the desired position and desired velocity are as follows: ; in, Indicates aircraft The expected position-time function; Indicates aircraft The expected velocity-time function;

[0168] (viii) Determination of acceleration commands for one's own aircraft at each moment during coordinated strikes Basic information about the aircraft The aircraft's own dynamic model is as follows: ; ; in, Indicates aircraft The velocity component along the x-axis at the current moment; Indicates aircraft The magnitude of the velocity at the current moment; Indicates aircraft The current flight path angle; Indicates aircraft Current heading angle; Indicates aircraft The velocity component along the y-axis at the current moment; Indicates aircraft The velocity component along the z-axis at the current moment; Indicates aircraft The rate of change of velocity at the current moment; Indicates aircraft The current axial overload command; Indicates aircraft The drag acceleration at the current moment; Represents gravitational acceleration; Indicates aircraft The rate of change of the current track angle; Indicates aircraft The first normal overload instruction at the current moment; Indicates aircraft The rate of change of the yaw angle at the current moment; Indicates aircraft The second normal overload instruction at the current moment.

[0169] The aircraft's own dynamics model can be represented as: ; in, Indicates aircraft The derivative of the position at the current moment; Indicates aircraft Current aircraft speed information; Indicates aircraft Current velocity derivative; Indicates aircraft Current time-instance perturbation vector representation; Indicates aircraft Rotation matrix at the current moment; aircraft The quality; Indicates aircraft The control quantity at the current moment; The state vector is represented as: ; Indicates aircraft Formation flight control commands; Rotation matrix Represented as: ; perturbation vector Represented as: .

[0170] When the flight mission is a formation control command, the following steps are performed for any aircraft within the mission group: Based on the acquired desired position and velocity commands of its own aircraft, as well as the desired position and velocity commands output by other aircraft in the mission group, the acceleration command of its own aircraft at each moment is determined during formation control. Specifically, this includes the following steps: When the number of aircraft in the mission group is equal to 1, the formula for calculating formation flight control commands is: ; in, Indicates aircraft Formation flight control commands; It represents a positive constant; Indicates aircraft The current position of the spacecraft; Show aircraft The desired position at the current moment; Represents another positive constant; Indicates aircraft The current speed of the aircraft; Indicates aircraft Expected velocity at the current moment; This refers to one of the aircraft belonging to the mission team.

[0171] When a task group performing a formation flight mission has more than one aircraft, and the task group is executing formation configuration operations, an aircraft is randomly selected as the leader aircraft. A cooperative control strategy is calculated and designed using formation flight control commands to ensure that the other aircraft follow the leader aircraft, maintaining the required geometric formation while synchronizing their movement with the leader aircraft. The leader aircraft transmits its position information to the other aircraft.

[0172] Synchronization with the motion of the aircraft can be expressed as: ; In the formula, Indicates aircraft Current position; This indicates the current position of the leading spacecraft. Indicates aircraft The formation configuration information.

[0173] At this point, the formula for calculating formation flight control commands is: ; in, Indicates aircraft Formation flight control commands; It represents a positive constant; Indicates aircraft The current position of the spacecraft; Represents another positive constant; Indicates aircraft The current speed of the aircraft; Indicates aircraft Expected velocity at the current moment; This refers to one of the aircraft belonging to the mission team.

[0174] When the number of aircraft in a mission group performing a formation flight mission is greater than one, and the mission group is performing a command tracking task for coordinated trajectory planning, then the control commands for each aircraft are designed as follows: ; in, Indicates aircraft Formation flight control commands; It represents a positive constant; Indicates aircraft The current position of the spacecraft; Indicates aircraft The desired position; Represents another positive constant; Indicates aircraft The current speed of the aircraft; Indicates aircraft The expected speed; This refers to one of the aircraft belonging to the mission team.

[0175] IX. Coordinated Strike Missions When the flight mission is a coordinated strike command, based on the desired position and velocity commands of the aircraft, the remaining time for the aircraft to reach the target, and the coordinated guidance acceleration calculation formula, the acceleration command of the aircraft at each moment during the coordinated strike is determined, and the following steps are performed for any aircraft in the mission group: Step 901, Obtaining Basic Information: Based on the acquired state information of its own aircraft at the current moment, the absolute value of the aircraft's speed at the current moment is used as the constant speed during the coordinated strike mission.

[0176] Based on the current state information of the aircraft and the current basic information of the target corresponding to the mission, as well as the position and coordinated trajectory of the aircraft, the instantaneous relative distance between the aircraft and the target is obtained.

[0177] Based on the current state information of the aircraft, the flight azimuth and pitch angles of the aircraft and the target line of sight in the inertial coordinate system are obtained, as well as the aircraft velocity azimuth and pitch angles of the aircraft relative to the line of sight coordinate system. Based on the basic information of the target at the current moment corresponding to the task, the target velocity azimuth and pitch angles relative to the line-of-sight coordinate system are obtained.

[0178] Step 902, Calculation of remaining flight time for the aircraft: The remaining flight time (TGO) of an aircraft is a core state quantity for time coordination, and its accurate estimation directly affects the effectiveness of saturation strikes and time-coordinated strikes.

[0179] Based on the instantaneous relative distance between the aircraft and the corresponding target, the aircraft's velocity pitch angle, and the formula for calculating the aircraft's remaining flight time, the remaining flight time of the aircraft is obtained and output.

[0180] Based on the relative motion characteristics between the aircraft and the target, the formula for calculating the remaining flight time of the aircraft is as follows: ; in, Indicates aircraft The remaining flight time; Indicates aircraft The instantaneous relative distance to the target; Indicates aircraft Constant velocity during coordinated guidance; Indicates aircraft The aircraft's velocity pitch angle relative to the line-of-sight coordinate system; The base navigation ratio is between 3 and 6.

[0181] Step 903, Calculation of the aircraft's desired strike command: Based on the remaining flight time of all aircraft in the task group and the formula for remaining time coordination deviation, the remaining time coordination deviation of the aircraft itself is obtained.

[0182] The formula for the remaining time coordination deviation is: ; in, Indicates aircraft Remaining time coordination deviation; This represents the group of aircraft in a mission group that performs a coordinated guidance mission. Indicates aircraft Constant velocity during coordinated guidance; It represents a positive number greater than 0.

[0183] Based on the remaining time coordination deviation of the aircraft, its azimuth angle, pitch angle, velocity azimuth angle, velocity pitch angle, and the calculation formula for the desired strike command, the desired strike control command of the aircraft is obtained. The desired strike control command includes the strike control axial overload command, the strike control first normal overload command, and the strike control second normal overload command.

[0184] The formula for calculating the desired strike command of an aircraft is: ; ; ; ; ; in, Indicates variable coefficient gain. It represents a positive constant; Represents a positive parameter; Represents a positive parameter; Represent a limiting function; Indicates aircraft The current axial overload command; Indicates aircraft The first normal overload instruction at the current moment; Indicates aircraft A constant speed; Indicates aircraft The time rate of change of the aircraft's azimuth angle; Indicates aircraft The aircraft's speed and pitch angle; Indicates aircraft The second normal overload instruction at the current moment; Indicates aircraft The time rate of change of the aircraft's azimuth angle; Indicates aircraft The aircraft's speed and azimuth angle; Indicates aircraft The azimuth angle of the aircraft; Indicates the desired angle of attack; Indicates aircraft The pitch angle of the aircraft.

[0185] This application proposes a feasible and implementable IOODA architecture working method and system for intelligent collaboration in aircraft swarms. It systematically presents the generalized logical flow of collaborative mission execution by aircraft swarms, clearly depicting the entire signal flow process from information perception and interaction processing to collaborative decision-making and action, achieving deep integration of the information interaction stage and the OODA loop. Furthermore, addressing the more complex and dynamic task scenarios that future swarm systems may face, this application further proposes intelligent management and control methods covering all aspects of IOODA, including mechanisms for dynamic task allocation, autonomous collaborative decision-making, distributed information fusion, and real-time adaptive adjustment. This supports swarms in achieving efficient, reliable, and self-organizing collaborative operation capabilities with minimal human intervention.

[0186] In the target collaborative recognition stage, the target recognition effect is enhanced through information interaction and positive / negative sample set enhancement methods. Simultaneously, a novel similarity discrimination criterion is constructed. The identity recognition program enables accurate classification and determination of targets.

[0187] The multi-sensor high-precision collaborative positioning and tracking algorithm improves the tracking and positioning accuracy of targets through a novel probability density weighted fusion method.

[0188] In the threat assessment stage, an LSTM-Transformer hybrid model is employed, using a deep neural network architecture for hierarchical intelligent online correction of target threat levels. Through threat level correction factor models, the autonomous evolution of the target group leads to the correction of collaborative adversarial behavior, resulting in dynamic threat assessment results.

[0189] In the target collaborative allocation stage, a distributed negotiation auction algorithm is proposed. Through continuous iteration of local revenue calculation and information interaction with neighbors (other aircraft in the group), the cluster will eventually converge to an allocation result state.

[0190] In the collaborative trajectory planning stage, considering the existence of certain planning errors and wind interference errors, an integrated planning and control method is introduced, and a collaborative trajectory planning iterative method based on time error is proposed, which can ensure high-precision planning results under multi-source interference.

[0191] In collaborative strike missions, an innovative collaborative guidance method based on maximum consistency error was designed by calculating the expected strike command of the aircraft (collaborative guidance algorithm), which can improve the time consistency accuracy during collaborative strikes.

[0192] Based on the same inventive concept, this application also provides a system based on IOODA control, which includes multiple aircraft. The solution provided by this system is similar to the solution described in the above method. Therefore, the specific limitations in the aircraft cluster embodiments provided below can be found in the limitations of the IOODA architecture working method for cluster intelligent collaboration described above.

[0193] The foregoing embodiments detailed key algorithms in cluster collaborative operations, including target collaborative identification, threat level assessment, task allocation, and trajectory planning. However, in the face of large-scale, highly dynamic cluster combat scenarios, the key to improving the intelligence level of cluster systems lies in how to organically integrate these algorithm modules and achieve autonomous and intelligent flow from information perception to collaborative execution based on changes in the battlefield situation. To this end, this application further proposes an "IOODA Brain" distributed state machine logic operation mechanism, integrating the aforementioned algorithms into a closed-loop framework of "information-observation-judgment-decision-action" to achieve intelligent management and control of cluster collaborative operations. In an exemplary embodiment, a system based on IOODA control, i.e., an aircraft cluster, implements the IOODA architecture working method for cluster intelligent collaboration described above through the following stages, i.e., the IOODA Brain distributed state machine logic operation principle, as follows: Figure 4 As shown.

[0194] Step 1001, Preparation Stage: The aircraft cluster includes a central control unit, an ad hoc network communication module, and multiple aircraft. Each aircraft includes a positioning module, a distance sensor module, an IOODA integrated module, a microcontroller control module, and a sensing module.

[0195] The IOODA integrated module includes an ad hoc network communication module. The aircraft establishes a communication and interaction architecture through the ad hoc network communication module.

[0196] Specifically, this application constructs a scenario considering multiple threat targets within a mission area. An aircraft swarm, comprising at least two aircraft, flies towards the mission area to perform a fully autonomous cooperative detection-strike mission, achieving a fully autonomous kill chain closure. Ground personnel transmit the target location information within the mission area to the IOODA brain mounted on each aircraft. Each IOODA brain runs the same target area segmentation algorithm, allocating different aircraft based on the size of the target segment and generating cooperative search tracks.

[0197] Ground personnel transmit the target location information to each aircraft's IOODA integrated module. Each aircraft's IOODA integrated module runs the same target area segmentation algorithm, assigns different aircraft according to the size of the target segment, and generates cooperative search tracks.

[0198] Step 1002, Self-organizing information exchange stage "I": Each aircraft forms an interactive network through a self-organizing network module. The network structure is a distributed, decentralized architecture. Each aircraft can communicate with any other aircraft as needed. The specific communication content and interaction nodes are implemented according to the working method described above.

[0199] Step 1002, Perception Stage "O": After the flight mission is assigned, each aircraft is equipped with an IOODA integrated module to send instructions to the aircraft to conduct the first search and detection along the predetermined cooperative search track. During the search, the detected targets (targets in the field of view) are photographed to obtain the observation images of the targets. Based on the observation images, the targets are pre-identified and the information is stored.

[0200] All aircraft IOODA integrated modules in each group interact via the self-organizing network master communication module ("I"), exchanging the target identification results stored in all aircraft. Each aircraft IOODA integrated module performs collaborative target identification and target enhancement identification, eliminates duplicate targets, and outputs the number of targets and the final category.

[0201] All aircraft in each group will conduct a second search along the collaborative search track, with each aircraft recording the relative distance, pitch angle, and azimuth angle of all targets.

[0202] Each group's IOODA integrated module interacts with all the target observation information stored by the aircraft through the self-organizing network general communication module. Each IOODA brain runs the target collaborative association algorithm, exchanges and determines coordinate information, and obtains information such as the rough position of the target.

[0203] Step 1003, Cognitive Stage "O": Each group's IOODA integrated module interacts with the self-organizing network's central communication module ("I"), exchanging target type and rough location information stored by all groups and all aircraft. Each aircraft's IOODA integrated module independently operates a threat level assessment matrix.

[0204] Step 1004, Collaborative Decision "D": Randomly select an aircraft as a central node, run the target collaborative allocation algorithm and the task collaborative allocation algorithm to obtain the target group serving each target and the task group executing each task, and distribute them to each aircraft through the interaction module "I".

[0205] Each aircraft forms a task group according to the mission coordination assignment instructions. Within each task group, one aircraft is randomly selected to run the mission formation configuration algorithm and the cooperative trajectory planning algorithm.

[0206] Step 1005, Coordinate the execution of "A": Each aircraft, according to the mission assignment instructions, interacts with other aircraft in the same group through the interaction module "I" to obtain the target's observation information and runs the multi-target, multi-sensor collaborative fusion tracking algorithm in the collaborative perception "O" to obtain the target's precise position and velocity information.

[0207] Based on the target information obtained through multi-vehicle collaboration, the next step is to execute the mission.

[0208] If it is a formation flight mission command, then the formation flight algorithm will be executed.

[0209] If it is a coordinated strike mission command, then the coordinated strike (coordinated guidance) algorithm will be executed.

[0210] If it is a collaborative reconnaissance mission command, then the collaborative trajectory-tracking control algorithm will be executed.

[0211] If the mission command is a collaborative detection / dense formation jamming / cover mission, then the formation control algorithm will be executed.

[0212] During each mission, the aircraft uploads images and status information of all aircraft to the ground station through the interactive module "I".

[0213] Step 1006, Evaluation and Iteration: If all aircraft execute the coordinated strike command, all aircraft will strike the target, and the entire mission will terminate.

[0214] If there is only one aircraft, repeat steps 1001 to 1004 above, except for step "I", until there are no aircraft left.

[0215] If there are more than two aircraft, repeat steps 1001 to 1004 above until there are no aircraft left.

[0216] The aforementioned IOODA Brain distributed state machine logic implements the flow of each stage through predefined "event-condition-action" rules, providing a clear framework for cluster collaboration. However, when the cluster scale expands and the task scenario becomes highly dynamic, relying solely on preset rules may be insufficient to handle all unexpected situations, and there is still room for improvement in the real-time performance and adaptability of its decision-making. To further enhance the intelligent decision-making capabilities of the IOODA Brain in complex environments and resolve the core contradiction between computing power allocation and real-time decision-making in large-scale cluster collaboration, this application also provides an IOODA Brain logic intelligent operation management system based on neural networks to replace the traditional "event-condition-action" rules, achieving smooth, learnable, and multi-factor decision-making state transitions. In an exemplary embodiment, addressing the problem of limited computing power and real-time decision-making in large-scale aircraft cluster collaborative task scenarios, this application also provides an IOODA Brain logic intelligent operation management system based on neural networks to replace the traditional "event-condition-action" rules, achieving smooth, learnable, and multi-factor decision-making state transitions. This system, as the intelligent decision-making core of the IOODA integrated module, has the following composition structure and operating principle: (I) Composition and Structure of the IOODA Brain Logic Intelligent Operation Management System like Figure 7 As shown, the IOODA Brain Logic Intelligent Operation Management System includes the following five core modules: The transition condition evaluation network is a multi-task learning network that serves as a global performance evaluation network. It combines the current state of the aircraft swarm system (obtained through a dynamic memory network) and its future state (obtained through a future prediction module) to evaluate task completion efficiency indicators. Specific indicators include the overall task completion rate and the urgency of a particular task in the next time step.

[0217] Intelligent Flow Decision Network: A pointer-like network with an attention mechanism for online optimization of the "structure" and "policy" of the state machine itself. The network receives current state features, predicted future state features, and transition condition evaluation values, calculates the weight distribution pointing to the next state, and supports smooth transitions to non-adjacent states.

[0218] Dynamic memory network: continuously reads and writes a memory matrix consisting of system events, state snapshots, and decision results, providing long-term dependencies for state transitions and supporting the accumulation and reuse of historical experience.

[0219] Future prediction module: A Transformer neural network module that can extrapolate the trajectory and mission completion status of the aircraft cluster based on the current state and the algorithm operation rules of each stage of IOODA, and generate a future state representation.

[0220] Perception encoder: After the individual aircraft in the cluster share information through the self-organizing network communication module (i.e., the "I" link), the perception encoder encodes the shared information to generate a unified state representation for use by other modules.

[0221] (II) Training and Operation Process of the IOODA Brain Logic Intelligence Management System Initialization: The system loads the initial state S0, initializes all neural network parameters, and clears the working memory of the dynamic memory network.

[0222] State activation and execution: Activate the neural network cluster corresponding to the current state node.

[0223] The current state node's perception encoder receives the current state information of the aircraft cluster and generates a state representation.

[0224] The future prediction module of the current state node represents the future state and drives business execution based on the current state and the algorithm operation rules of each stage of IOODA.

[0225] The dynamic memory network of the current state node continuously accumulates state information over a period of time.

[0226] Parallel transfer evaluation: The transfer condition evaluation network continuously monitors the performance indicators of the task completion efficiency and outputs the transfer condition evaluation value.

[0227] Intelligent transfer decision-making: The intelligent transfer decision-making network calculates the optimal transfer event based on the characteristics of the current state, the predicted future state, and the evaluation value of the transfer conditions. The optimal transfer event includes the next stage IOODA mission instruction and the aircraft number that will execute the mission.

[0228] State switching and updating: During offline training, the decision to perform a state transition is based on the experience of combat experts. If a transition is decided, the system smoothly switches to the target state and stores the transition (source state, target state, transition effectiveness) as an experience point in the dynamic memory network. All relevant neural networks update their parameters through online learning based on memory playback.

[0229] During online operation, the system directly receives the current state characteristics, the predicted future state characteristics, and the evaluation value of the transition conditions. Through the instructions output by the intelligent flow decision network, it points to the weight distribution of the next state, supporting smooth transitions to non-adjacent states.

[0230] Through the aforementioned IOODA Brain Logic Intelligent Operation Management System, the aircraft swarm can achieve intelligent management throughout the entire lifecycle, from information perception and interactive processing to collaborative decision-making and action, with minimal human intervention, further enhancing the swarm's adaptability and decision-making efficiency in complex and dynamic mission scenarios.

[0231] In one exemplary embodiment, such as Figure 5 As shown, an IOODA brain hardware structure is provided to support the operation of the above-mentioned distributed state machine logic.

[0232] In one exemplary embodiment, such as Figure 6 As shown, an IOODA brain-distributed state machine and hardware structure are provided. This structure defines the entire process of state transition mechanism of the aircraft cluster from information interaction (I), perception (O), cognition (O), decision-making (D) to execution (A), and clarifies the hardware composition of each aircraft's IOODA integrated module, including self-organizing network communication module, positioning module, perception module, microcontroller control module, etc., providing integrated software and hardware support for subsequent collaborative working methods.

[0233] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties. Moreover, the collection, use and processing of the relevant data are carried out in compliance with the relevant data protection laws and policies of the country where the location is located, and with the authorization granted by the owner of the corresponding device.

[0234] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0235] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0236] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0237] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A working method for an IOODA architecture with intelligent cluster collaboration, characterized in that, The cluster comprises multiple aircraft; each aircraft is used to execute the IOODA architecture working method for cluster intelligent collaboration; The working method of the IOODA architecture for intelligent cluster collaboration includes: The system acquires in real-time status information of all aircraft within a target area, as well as basic information of all targets within that area; the status information includes position and velocity information; the basic information includes status and category information. Based on the status information of its own aircraft and the basic information of all targets in the target area, the threat level of its own aircraft to each target is calculated. Based on the threat level of the aircraft to each target and the initial target allocation vectors output by all aircraft in the target area, the target allocation results of the aircraft are calculated and output. Based on the task cost function matrix corresponding to the target group and the provisional task allocation vector output by all aircraft in the target group, the task allocation result corresponding to its own aircraft is calculated and output; the target group is determined according to the target allocation result; all aircraft in the target group correspond to the same target; Based on the basic information of the target corresponding to the task group and the number of aircraft in the task group, the mission terminal location information corresponding to its own aircraft is determined, and based on the mission terminal location information corresponding to its own aircraft and the status information of all aircraft in the task group, the virtual flight time of its own aircraft is determined; the task group is determined according to the task allocation result, and all aircraft in the task group correspond to the same task. Based on the status information of its own aircraft within the task group, the location information of the corresponding task terminal of its own aircraft, and the virtual flight time of its own aircraft, the cooperative trajectory corresponding to its own aircraft within the task group is calculated, and the corresponding flight mission is executed based on the cooperative trajectory corresponding to its own aircraft.

2. The working method of the IOODA architecture for cluster intelligent collaboration according to claim 1, characterized in that, The method for obtaining the category information specifically includes: The system acquires a set of observation images of all targets detected within the target area, performs target recognition based on the set of observation images, and outputs preliminary recognition results of all targets identified by the aircraft. The preliminary recognition results include preliminary confidence, preliminary category information, and preliminary feature descriptors. Based on the preliminary identification results of all targets output by all aircraft corresponding to the target area, the similarity between the preliminary identification results of the first target and the preliminary identification results of the second target is calculated. When the similarity is greater than a first threshold, the first target is determined to be an independent target. When the similarity is less than or equal to the first threshold, the second target is determined to be a similar target of the first target, thus obtaining a set of similar targets of the first target. The first target is any target identified by the aircraft itself, and the second target is any target identified by other aircraft. Based on the set of similar targets to the first target and the preliminary identification results of the first target, the preliminary feature descriptor is corrected to obtain the corrected feature descriptor; Based on the modified feature descriptors, the category information of the corresponding target is determined.

3. The working method of the IOODA architecture for cluster intelligent collaboration according to claim 1, characterized in that, Based on the aircraft's own status information and the basic information of all targets within the target area, the threat level of the aircraft to each target is calculated, specifically including: Based on the status information of its own aircraft and the basic information of all targets in the target area, the threat element index of its own aircraft to each target is determined; the threat element index includes type threat index, speed threat index, distance threat index and angle threat index; The relative threat likelihood probability of the aircraft to each target is obtained by weighted summation of the threat factor indices. The initial relative threat probability of the aircraft to each target is randomly determined. Based on the relative threat likelihood probability and the initial relative threat probability, the initial threat level of the aircraft to each target is obtained through Bayes' formula. Using the flight time, flight status of the aircraft, and flight status of the target as inputs, a threat correction factor model is used to determine a correction factor, and the initial threat level of each target is corrected using the correction factor to obtain the threat level of the aircraft to each target; the flight status includes three-dimensional position and three-dimensional velocity.

4. The working method of the IOODA architecture for cluster intelligent collaboration according to claim 1, characterized in that, Based on the acquired threat level of the aircraft to each target and the initial target allocation vectors output by all aircraft corresponding to the target area, the target allocation results of the aircraft are calculated, specifically including: Randomly determine and output the initial target assignment vector for your own aircraft; Obtain the initial target allocation vectors for other aircraft corresponding to the target area; The results of the initial target allocation vectors for the same target are summed to obtain the target conflict perception of the aircraft for each target. Based on the aircraft's perception of target conflict with each target and the threat level of each target to each target, calculate the aircraft's overall target benefit for each target. Select the target corresponding to the maximum comprehensive benefit as the provisional target of your own aircraft, and calculate and output the provisional target allocation vector of your own aircraft; Obtain the provisional target allocation vectors of other aircraft corresponding to the target area. Based on the provisional target allocation vectors of all aircraft corresponding to the target area, determine whether the provisional target allocation vectors of other aircraft are inconsistent with the provisional target allocation vectors of the aircraft itself. If so, then use the provisional target allocation vector of its own aircraft as the initial target allocation vector of its own aircraft, and return to the step "Get the initial target allocation vector of other aircraft corresponding to the target area"; If not, then the provisional target of the aircraft itself will be taken as the final target of the aircraft itself, and the target allocation result of the aircraft itself will be obtained.

5. The working method of the IOODA architecture for cluster intelligent collaboration according to claim 1, characterized in that, Based on the basic information of the target corresponding to the mission group and the number of aircraft within the mission group, determine the mission terminal location information corresponding to your own aircraft, specifically including: When the number of aircraft in the task group is equal to 1, the position information of the target corresponding to the task group is used as the position information of the task terminal. When the number of aircraft in a mission group is greater than 1, and the aircraft in the mission group fly in formation throughout the entire process, the position information of the target corresponding to the mission group is used as the mission terminal position information. When there is more than one aircraft in the mission group and there is a target, the aircraft in the mission group perform formation configuration operations to determine the formation configuration information of each aircraft in the mission group. Based on the formation configuration information of each aircraft and the basic information of the target corresponding to the mission group, the mission terminal position information of each aircraft in the mission group is obtained.

6. The working method of the IOODA architecture for cluster intelligent collaboration according to claim 5, characterized in that, Determine the formation configuration information for each aircraft within the mission group, specifically including: Randomly select one aircraft from the mission team to perform the following operations: Based on the acquired task type elements, the overall fitness of the formation corresponding to the task is obtained; Based on the acquired status information of all aircraft in the mission group, the position center of each aircraft in the mission group is calculated, and based on the position center of each aircraft in the mission group, the mission configuration center is calculated. Randomly initialize the formation parameters of each aircraft in the task group to obtain the initial formation parameters of each aircraft in the task group; Based on the mission configuration center, the initial formation parameters of each aircraft in the mission group, and the gradient descent aircraft formation optimization algorithm, the expected relative elevation angle, expected relative yaw angle, and expected relative distance of each aircraft in the mission group to the mission configuration center are obtained. Based on the expected relative elevation angle, expected relative yaw angle, and expected relative distance of each aircraft in the mission group to the mission configuration center, as well as the formation configuration calculation formula, the formation configuration information of each aircraft in the mission group is calculated and output.

7. The working method of the IOODA architecture for cluster intelligent collaboration according to claim 1, characterized in that, The coordinated trajectory includes a position coordinated trajectory and a velocity coordinated trajectory; the position coordinated trajectory includes the desired position command at each moment; the velocity coordinated trajectory includes the desired velocity command at each moment; based on the status information of the aircraft within the task group, the position information of the corresponding mission terminal of the aircraft, and the virtual flight time of the aircraft, the coordinated trajectory corresponding to the aircraft within the task group is calculated, specifically including: Based on the status information of its own aircraft within the task group, the location information of the task terminal corresponding to its own aircraft, and the virtual flight time of its own aircraft, the position coordination trajectory and the preliminary velocity coordination trajectory are obtained. Based on the virtual flight time of its own aircraft, the initial speed coordination trajectory is corrected to obtain the speed coordination trajectory; Based on the position-coordinated trajectory and velocity-coordinated trajectory, the desired position and velocity commands of each aircraft within the mission group at each moment are determined and output.

8. The working method of the IOODA architecture for cluster intelligent collaboration according to claim 1, characterized in that, The cooperative trajectory includes a position cooperative trajectory and a velocity cooperative trajectory; the position cooperative trajectory includes the expected position command at each moment; the velocity cooperative trajectory includes the expected velocity command at each moment; and further includes: when the flight mission is a formation control command, determining the acceleration command of the aircraft at each moment during the formation control process based on the expected position command and expected velocity command of the aircraft itself, as well as the expected position command and expected velocity command output by other aircraft in the mission group.

9. The working method of the IOODA architecture for cluster intelligent collaboration according to claim 1, characterized in that, The coordinated trajectory includes a position coordinated trajectory and a velocity coordinated trajectory; the position coordinated trajectory includes the expected position command at each moment; the velocity coordinated trajectory includes the expected velocity command at each moment; and further includes: when the flight mission is a coordinated strike command, determining the acceleration command of the own aircraft at each moment during the coordinated strike based on the expected position command and expected velocity command of the own aircraft, the remaining time for the own aircraft to reach the target, and the coordinated guidance acceleration calculation formula.

10. A system based on IOODA control, characterized in that, The IOODA-based control system includes multiple aircraft; each aircraft is used to execute the IOODA architecture working method of cluster intelligent collaboration as described in any one of claims 1-9.

Images