Aircraft collaborative task permission allocation method and equipment based on multi-attribute decision

Abstract

The invention discloses an aircraft collaborative task permission allocation method and equipment based on multi-attribute decision, and belongs to the technical field of aircrafts. The method comprises the following steps: S1, hierarchical design of an auxiliary system: dividing gradient autonomous levels, determining responsibility boundaries of operator-auxiliary system-aircraft under different levels, and defining each level as a scheme to form a scheme set; s2, target attribute modeling: converting key factors influencing permission allocation into computable quantitative indexes; s3, establishing a decision matrix: converting the quantification result into a corresponding relation of authority level-attribute cognitive quality to obtain the decision matrix; s4, determining a weight coefficient: determining evaluation indexes and distributing a reasonable weight for each evaluation index; and S5, solving authority distribution levels: based on the decision matrix and the weight, calculating a comprehensive score of each authority level, and determining an optimal level after sorting. According to the invention, the efficiency and safety of man-machine cooperation can be improved, and the flexibility and adaptability of the system are enhanced.

Description

Technical Field

[0001] This invention belongs to the field of aircraft technology, specifically relating to a method and device for allocating permissions for collaborative tasks in aircraft based on multi-attribute decision-making. Background Technology

[0002] The core value of a human-machine integrated mission system lies in improving mission execution efficiency through the complementary advantages of humans and machines. However, this system faces significant bottlenecks in practical applications: constrained by airborne hardware resources, payload capacity, and energy consumption, the ratio of human-machine command to aircraft control by operators is far lower than that of ground station command models, making it difficult to achieve efficient management and control of large-scale aircraft swarms.

[0003] In complex mission environments, situations change rapidly. Operators need to simultaneously undertake multiple tasks such as situation assessment, instruction issuance, formation coordination, and emergency response. The contradiction between the low human-machine ratio and the high-intensity mission requirements directly leads to a sharp increase in the workload of operators, far exceeding the physiological and cognitive load limits of the human body, which can easily cause problems such as decision-making delays and operational errors.

[0004] The current collaborative model faces a dilemma of polarization: if it relies too much on human decision-making, it will not only be difficult to cope with complex and dynamic environments, but also lead to low collaborative efficiency due to human fatigue and information overload; if it relies solely on autonomous systems, it will be limited by the maturity of the technology, resulting in problems such as poor system adaptability to sudden scenarios, insufficient decision credibility, and weak fault tolerance, which will seriously affect the smooth progress of the task.

[0005] Therefore, achieving dynamic matching and precise adaptation of human-machine system capabilities has become the key to overcoming the above-mentioned dilemmas. Summary of the Invention

[0006] The present invention aims to at least partially solve one of the technical problems in the aforementioned related technologies.

[0007] Therefore, the purpose of this invention is to provide a method and device for allocating permissions for collaborative missions of aircraft based on multi-attribute decision-making. It specifically designs a method for classifying the autonomous level and allocating permissions in human / machine collaborative systems, which can balance the boundaries between human intervention and autonomous decision-making through a scientific human-machine permission adaptation mechanism, and provide technical support for improving the stability, efficiency and reliability of collaborative missions.

[0008] To solve the above-mentioned technical problems, the present invention is implemented as follows: This invention provides a method for allocating permissions for collaborative missions of aircraft based on multi-attribute decision-making, the method comprising: S1. Hierarchical Design of Auxiliary Systems: Divide the autonomous levels into tiered levels, clarify the responsibilities of operators, auxiliary systems and aircraft at different levels, define each level as a solution, and form a solution set; S2. Target attribute modeling: Transform the key factors affecting permission allocation into calculable quantitative indicators to avoid subjective decision-making; S3. Establish a decision matrix: Transform the quantitative results into a correspondence between permission level and attribute cognitive quality to obtain a decision matrix, providing a data carrier for subsequent decision-making; S4. Determine the weighting coefficients: Determine the evaluation indicators and assign reasonable weights to each evaluation indicator to reflect the differences in the importance of the indicators; S5. Solve for permission allocation levels: Based on the decision matrix and weights, calculate the comprehensive score of each permission level, sort them, and determine the optimal level.

[0009] In addition, the aircraft cooperative task permission allocation method based on multi-attribute decision-making according to the present invention may also have the following additional technical features: In some implementations, the levels in step S1 include auxiliary control level, semi-autonomous level, highly autonomous level, and fully autonomous level.

[0010] In some implementations, the operator's responsibilities in the auxiliary control level include: issuing flight maneuver and detection mode commands, and assessing the situation; the auxiliary system functions include: generating control data, monitoring flight status, and marking sensor anomalies; and the aircraft's capabilities include: executing commands and basic obstacle avoidance. In the semi-autonomous stage, the operator's responsibilities are: setting the route and detection mode, and selecting the path optimization scheme; the auxiliary system functions are: automatically generating control data, optimizing the formation path, and marking high-value targets; the aircraft's capabilities are: executing the approved path and optimizing sensor allocation; The operator's responsibilities in the highly autonomous level are: autonomously planning paths and adjusting formations, allocating sensor modes, and generating threat assessments; the auxiliary system functions are: autonomous coordinated actions and inter-aircraft data synchronization; the aircraft's capabilities are: autonomous coordinated actions and inter-aircraft data synchronization. The operator's responsibilities in the fully autonomous class are: setting strategic objectives and intelligence requirements, and monitoring the credibility of data; the auxiliary system functions are: planning dynamic task chains, target identification, and reconstructing sensor networks; the aircraft's capabilities are: swarm intelligence and autonomous adaptation to complex environments.

[0011] In some implementations, step S2 includes: S21. Multi-attribute decision problem modeling: Solve the priority ranking or evaluation problem of several feasible solutions under multiple objectives by solving the complete order of feasible solutions; S22. Task Attribute Modeling: Decompose task attribute information into task objective difficulty and task execution status; establish a task objective difficulty evaluation model for task objective difficulty and a task objective achievement evaluation model for task execution status to evaluate the current task execution completion level; S23. Environmental Attribute Modeling: Divide the left and right sides of the aircraft into fan-shaped regions and construct a threat level model to quantify the threat level of each fan-shaped region to the aircraft; S24. Aircraft capability modeling: Quantify the aircraft's detection capabilities and anti-jamming capabilities separately; S25. Operator Command Decision Attribute Modeling: Quantitatively model command decision preferences using survival priority coefficients, and introduce decision load coefficients as performance evaluation indicators to model auxiliary decision-making capabilities.

[0012] In some of these implementations, the detection capability of the aircraft in S24 is quantified using coverage factor, overlap factor, and jamming suppression ratio; The anti-jamming capability is quantified using antenna gain, radar average power, signal bandwidth, signal illumination time, target reflective area, and radar resolution volume unit.

[0013] In some implementations, step S3 includes: Matrix element definition: Let x be the cognitive quality of the i-th attribute at the j-th permission level. ij Cognitive quality x ij This indicates the aircraft's level of awareness of attribute i at level j; Introducing a utility function: linking the probability of user intervention with the quality of attribute cognition, and ensuring that the quality of cognition can be quantified into decision utility through a pre-defined formula.

[0014] In some implementations, step S4 includes: S41. Provide a detailed description of the difficulty of mission completion, the degree of mission objective completion, the degree of environmental threat, the degree of space threat, detection capability, anti-jamming capability, command and decision-making capability, and auxiliary decision-making capability, and obtain evaluation values ​​for the environment, mission, user, and aircraft capabilities; S42. Multiple performance evaluation indicators: Evaluation functions are used to evaluate each indicator to obtain a decision matrix; S43. Calculate the optimal permission allocation level based on the human-machine permission allocation level adjustment algorithm.

[0015] In some implementations, step S43 uses a comprehensive weighted method in group decision-making to solve for the permission allocation level.

[0016] In some of these embodiments, step S43 includes: Determine the weights of different decision attributes; Calculate the priority value for each pair of permission assignment levels; The priority order of the current permission allocation level is obtained based on the priority value; The evaluation matrix is ​​composed of the priority order calculated from multiple aircraft in the cluster.

[0017] This invention also provides a computer device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the steps of the aircraft cooperative mission permission allocation method based on multi-attribute decision-making as described in any of the preceding embodiments.

[0018] Compared with the prior art, the present invention has at least the following beneficial effects: In this embodiment of the invention, the provided method for allocating permissions for collaborative aircraft tasks based on multi-attribute decision-making clearly designs the autonomy level of the aircraft and the autonomy level of the human / machine collaborative task system. This method comprehensively considers multiple attributes such as task completion difficulty, task objective completion degree, environmental threat level, space threat level, detection capability, anti-interference capability, operator command and decision-making capability, and auxiliary decision-making capability as inputs to the decision-making algorithm. By constructing a decision matrix and determining attribute weights, it achieves accurate calculation of the human-machine permission allocation level and assigns the task to the most suitable executor. This method not only improves the efficiency and safety of human-machine collaborative tasks but also enhances the flexibility and adaptability of the system.

[0019] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0020] Figure 1 This is a structural diagram of the evaluation index disclosed in one embodiment of the present invention; Figure 2 This is a decision evaluation matrix diagram for multi-aircraft permission allocation levels disclosed in one embodiment of the present invention. Detailed Implementation

[0021] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] The embodiments of the present invention will be described in detail below with reference to the accompanying drawings and specific examples and application scenarios.

[0023] This invention provides a method for allocating permissions for collaborative aircraft tasks based on multi-attribute decision-making. By decomposing key process elements, it clearly designs the autonomy levels of the aircraft and the human / machine collaborative task system. This method comprehensively considers multiple attributes as inputs to the decision-making algorithm, including task completion difficulty, task objective completion rate, environmental threat level, space threat level, detection capability, anti-interference capability, operator command and decision-making ability, and auxiliary decision-making capability. By constructing a decision matrix and determining attribute weights, it achieves precise calculation of the human-machine permission allocation level, assigning tasks to the most suitable executor. This method not only improves the efficiency and safety of human-machine collaborative tasks but also enhances the system's flexibility and adaptability.

[0024] Please see Figure 1 As shown, in some embodiments of the present invention, a method for allocating permissions for collaborative tasks between aircraft (operator / aircraft) based on multi-attribute decision-making is provided, including the following steps: Step 1: Hierarchical Design of Auxiliary System and Confirmation of Feasibility Solution Set Level 1: Auxiliary Control The operator issues flight maneuver commands and detection mode control commands, while simultaneously assessing the situation. The auxiliary system generates control data based on the received maneuvers and sensor patterns and transmits it to the aircraft. The system continuously monitors the aircraft's flight status, analyzes sensor data, and flags abnormal states.

[0025] Level 2: Semi-autonomous The operator specifies the flight path and sets the cooperative detection mode. The auxiliary system automatically generates control data based on the detection mode. An optimized formation path is generated based on environmental threats and returned to the operator for selection.

[0026] Level 3: High Autonomy The operator only needs to set the flight mission and detection targets and supervise the process. The auxiliary system autonomously plans the flight path, adjusts the formation, avoids obstacles, and automatically allocates sensor cooperative detection modes. When an abnormal situation occurs, feedback is sent back to the operator for decision-making.

[0027] Level 4: Complete Autonomy Operators need to set strategic objectives and define intelligence requirement levels, and monitor data reliability. The auxiliary system will autonomously plan dynamic mission chains for multiple aircraft, optimize formation efficiency in real time, achieve intelligent target identification based on models, and dynamically reconstruct sensor network topology to achieve full autonomy.

[0028] The feasible solution set is a collection of autonomy levels defined according to different ranges of autonomy capabilities; that is, the feasible solution set is... .

[0029] Step 2: Target Attribute Modeling The evaluation of target attributes provides input data for autonomy level decision-making. It includes four parts: environment, mission, aircraft, and operator capabilities. It provides the cognitive quality of the system at each level for authority allocation level decision-making and is a key model that provides a basis for autonomy level decision-making.

[0030] Step 2.1: Environmental Attribute Modeling In this invention, the aircraft utilizes airborne radar to acquire information and determine the current environmental complexity. Based on the information acquired by the airborne radar, and to facilitate subsequent data processing, the aircraft is positioned 90 degrees to each of its left and right sides during flight. The area is divided into six sector areas, each occupying 30 square meters. The range, assuming This represents the sensitivity weight of the unmanned system to a sector in a certain direction. To facilitate subsequent data processing and ensure that the environmental threat level evaluation index has the same order of magnitude as other evaluation indices, the sensitivity weight vector is required to... satisfy:

[0031] In the formula, For vectors of Each component.

[0032] Furthermore, the distance between the air defense zone and the aircraft also affects the level of environmental threat to the cluster. This indicates the aircraft's spatial sensitivity and threat level to the air defense zone. It is a function of distance; the degree of danger is related to the distance between the air defense zone and the aircraft. They form a certain linear relationship, and the calculation formula is:

[0033] In the formula, Indicates the safe radius of the aircraft. This indicates the maximum detection radius of the aircraft.

[0034] Based on the threat level and sensitivity weights, the environmental threat level posed by a single air defense zone to an aircraft is calculated using the following formula:

[0035] In summary, the formula for calculating the environmental complexity of a single aircraft is as follows:

[0036] In the formula, Indicates the first The set of all air defense zones detected in a sector-shaped region.

[0037] Step 2.2: Task Attribute Modeling Task attribute information constitutes the primary input element in permission allocation level decisions. Its content can be divided into two parts: the difficulty of the current task objective and the task execution status. This can be represented as a set. ,in Represents the difficulty of the current task. This indicates the completion status of the task.

[0038] Mission objective difficulty assessment evaluates the difficulty of the mission being performed by the aircraft. The specific assessment method is as follows:

[0039] in, This represents the average difficulty of executing the current task; The size of the current aircraft formation is normalized. This represents the difficulty of the action type performed in the current task, where , Represents the weights of different information items. . The value is 0, which indicates that the current task is very difficult, and 1 indicates that the current task can be executed smoothly.

[0040] Mission objective achievement is an evaluation of the aircraft's performance in completing the current mission. The specific evaluation method is as follows:

[0041] in, It is time The distance traveled inwards, It is the distance between the two target points. It is the distance between the current moment and the next target point; This is the aircraft's current heading. It is the current desired heading planned by the aircraft, and it is the maximum difference between the real-time heading and the desired heading during the mission execution. This is the aircraft's current speed. It is the current expected speed of the aircraft's design. It is the maximum real-time speed of the aircraft during the mission. Represents the weights of different information items. . , where 0 represents that the current task is completely unexecutable, and 1 represents that the current task can be executed successfully.

[0042] Step 2.3: Aircraft Capability Modeling The detection range of multi-vehicle cooperative operation is the union of the detection ranges of all individual vehicles in the mission system. The influencing factor of the detection capability index is: coverage coefficient. , Interference suppression ratio Its calculation model is as follows:

[0043] In the above formula, , , Represents the weighting factor, and .

[0044] The coverage factor is expressed as the size of the area of ​​responsibility occupied by the system's detection area, i.e.:

[0045] The overlap coefficient is represented by the ratio of the sum of the detection areas of all detection equipment to the detection area of ​​the entire early warning detection system, that is:

[0046] In environments with electronic interference, the detection area of ​​an early warning detection system will decrease. The interference suppression ratio reflects the change in the detection area of ​​the early warning detection system after being interfered with, and its calculation model is as follows:

[0047] In the above formula, This indicates that after being interfered with, the first [unit / item] in the system... The detection area of ​​each detection device.

[0048] In environments with electronic interference, the effectiveness of radar anti-jamming capabilities directly impacts the early warning and detection performance of a mission system. Radar anti-jamming capability is reflected in three indicators: basic anti-jamming capability... Work system and its ability to resist interference Technical measures to resist interference The calculation model for radar anti-jamming capability is as follows:

[0049] In the above formula, Represents the adjustment factor. , , The weights represent the three anti-interference capabilities.

[0050] The basic anti-interference capability calculation model is as follows:

[0051] In the above formula, Indicates the average power of the radar. Represents antenna gain. and These represent the signal bandwidth and the signal illumination time, respectively. This represents the detectable reflective area of ​​the target. This represents the radar resolution volume unit.

[0052] The anti-interference capability of the work system is quantified using an expert scoring method, and its calculation model is as follows:

[0053] In the above formula, Depending on whether the radar used the first The different work systems are assigned 1 or 0 respectively. Indicates the first The contribution of each work system is shown in Table 1.1. Table 1.1 Contribution of Each Working System

[0054] The calculation model for the anti-interference capability of technical measures is as follows:

[0055] In the above formula, Depending on whether the radar used the first The anti-interference measures are respectively set to 1 or 0. Indicates the first The contribution of each anti-interference measure is shown in Table 1.2.

[0056] Table 1.2 Contribution of Each Working System

[0057] Step 2.4: Operator Command Decision Attribute Modeling Step 2.4.1: Command and Decision-Making Capabilities Command decision preferences are expressed through a survival priority coefficient. Quantitative modeling is performed in three typical scenarios: When the value equals 0, the system adopts an extreme offensive strategy, prioritizing the probability of target destruction without considering the platform's survivability. When the value is 1, the system executes a conservative task strategy, which requires that the human-machine system must be kept alive. At that time, decision-makers dynamically weigh the degree of task completion against the probability of survival. The closer the value is to 1, the higher the importance attached to system survival.

[0058] In this mathematical model, the survival-damage joint probability parameter Characterization within a preset time window Within, the cumulative probability measure of an aircraft simultaneously achieving survivability and target destruction effectiveness in a high-threat area:

[0059] This probability measure is defined as: Within the preset time frame Within a high-threat environment, the conditional probability of achieving target destruction without guaranteeing platform survival is as follows:

[0060] This mathematical model incorporates multi-source threat assessment parameters, where... , ,in This indicates the cumulative frequency of enemy detection and fire threats to a mission unit per unit of time. Defined as the average time from mission initiation to target contact, its numerical characteristics are jointly determined by the target search, localization and cooperative detection effectiveness, and are affected by sensor performance and system configuration; A measure of the probability that a task unit will complete an assault mission; This is the survival probability parameter for the task unit in a threat environment.

[0061] Step 2.4.2: Decision Support Capability To quantify the system's effectiveness in improving command load, a decision load coefficient is introduced. It serves as a performance evaluation indicator. Its mathematical representation model is as follows:

[0062] Step 3: Establish the decision matrix After obtaining the four evaluation values ​​from the attribute model in step 2, these four values ​​are used as inputs for the permission allocation level decision. Then, a decision matrix is ​​constructed based on multi-objective attribute decision theory. Let the first... Under each attribute, the permission level When, the matrix value is That is, the aircraft's level of authority allocation Next to the Cognitive quality of each attribute. Define utility function. Its definition is as follows, where, .

[0063]

[0064] in, It is the permission allocation level The probability of user intervention; For the aircraft's target attributes Cognitive quality score.

[0065] Step 4: Determine the weighting coefficients The human-machine permission allocation level decision-making process utilizes multi-objective attribute functions to obtain indicator attribute values ​​under different permission allocation levels. Then, a decision-making algorithm is used to calculate the human-machine permission allocation level, which consists of three steps: Step 4.1: Provide a detailed description of the difficulty of mission completion, the degree of mission objective completion, the level of environmental threat, the level of space threat, detection capability, anti-interference capability, command and decision-making capability, and auxiliary decision-making capability to obtain evaluation values ​​for the environment, mission, user, and aircraft capabilities; Step 4.2: Multiple performance evaluation metrics, using evaluation functions By evaluating each indicator, a decision matrix is ​​obtained. Step 4.3: Human-machine permission allocation level adjustment algorithm, calculate the optimal permission allocation level based on the algorithm.

[0066] The algorithm for adjusting the human-machine permission allocation level consists of the following parts. First, after obtaining the decision matrix using the decision matrix method, the values ​​in each attribute column are compared to determine the priority of different permission allocation levels. The calculation method is as follows:

[0067] and The values ​​in the decision matrix, The priority value is the permission assignment level on target k. about The priority of the function. A smaller function value indicates... and In the target k The difference is small; when When it is 0, it means and There is almost no difference; when Approaching 1, Compare High priority; when When equal to 1, the permission assignment level is... Strictly superior to ,when When it is negative, it means Superior After obtaining priority values, since different objectives have varying importance in the permission allocation level decision-making process, it is necessary to assign corresponding weights to each objective to differentiate their importance. That is, under different weights, the degree of importance given to objective optimization differs, and the vector... Represents the weights of different decision attributes. , The weights for mission completion difficulty, mission objective completion rate, environmental threat level, space threat level, detection capability, anti-interference capability, command and decision-making capability, and auxiliary decision-making capability are respectively given. The target importance values ​​are shown in Table 1.3.

[0068] Table 1.3 Target Importance Values

[0069] Define the target importance value table and define variables. The judgment matrix is ​​obtained. D According to the formula , , obtain the vector .

[0070]

[0071]

[0072] In the formula, I It is the identity matrix, if the target importance matrix D The value in the equation is accurately estimated, and the above equation is strictly equal to 0. D If the estimate is inaccurate, then D A perturbation with a small element means a perturbation with a small eigenvalue, as follows:

[0073] In the formula, It is a matrix D The maximum eigenvalue is obtained from the above formula, and the weight vector is obtained from it. This method is called the eigenvector method.

[0074] When determining weights using this method, one can use... To measure D Elements Consistency in estimation is obtained, and therefore a consistency evaluation index is introduced. CI :

[0075] In the formula, n To determine the order of a matrix. CI Random index of the same order matrix given in Table 1.4 RI The ratio is called the consistency ratio. CR ,Right now: .

[0076] Table 1.4 Average Random Consistency Index

[0077] when When the consistency of the judgment matrix is ​​considered acceptable, then the consistency of the judgment matrix is ​​acceptable. Conversely, when... In such cases, the judgment matrix should be appropriately modified to maintain a certain degree of consistency. For first- and second-order matrices, this holds true consistently, ensuring complete consistency. .

[0078] Step 5: Solve for permission allocation levels The steps for solving the permission allocation level using the comprehensive weighted method include: (1) Each decision-maker makes a multi-indicator decision on each option individually, assuming the first option is the first option. The aircraft to the first The first scheme The target evaluation value is The evaluation values ​​were determined using a standardized scoring system, resulting in an evaluation matrix. In the calculation process, this involves assigning priority levels to individual entities within a multi-aircraft fleet based on their current priority authority. .

[0079] (2) Calculate the arithmetic weighted average score and the geometric weighted average score for each scheme. The arithmetic weighted average score is... , For the first The weight values ​​of each objective. The geometrically weighted average is divided into... .

[0080] (3) Calculate the mixed average score for each scheme. .

[0081] (4) Calculate the overall average score for each scheme. .

[0082] (5) Calculate the evaluation coefficients for each scheme. .

[0083] (6) Based on the evaluation coefficient Arrange the priority levels of the permission allocation schemes from highest to lowest. .

[0084] In priority This represents the current optimal permission allocation level, enabling the adjustment of permission allocation levels in the human / machine collaborative system.

[0085] In the above embodiment, a priority sequence was obtained through six steps, thus selecting the optimal solution. In this solution, a level is assigned to each pair of permissions based on the weights of different decision attributes. The priority value is calculated using a formula, representing the degree of priority of the current autonomy level compared to the other three autonomy levels. This is done by calculating the priority value for each permission allocation level, i.e., the comparison. The size of the value indicates the relative merit of each permission level. If... Then it is defined as That is, the level of permission allocation Superior Based on the above process, each aircraft in the human-machine permission allocation under the current system autonomy level can obtain a priority ranking of its current permission allocation level. , This represents the aircraft's serial number during the mission execution process. Assign priority values ​​to permission levels.

[0086]

[0087] The range of values ​​is determined by prioritizing the access control levels of each aircraft. Subsequently, the authority allocation levels are solved using the comprehensive weighted method in group decision-making. The priority ranking calculated by multiple aircraft forms an evaluation matrix. The decision scheme set consists of four authority allocation levels, with each aircraft acting as the decision-maker. Each aircraft has a different weight coefficient based on its responsibilities in different tasks. The weighting coefficient for each aircraft is: , , .

[0088] Example 1: In this embodiment, the aircraft cooperative task permission allocation method based on multi-attribute decision-making includes the following steps: Step 1: Confirmation of Feasibility Study Set The feasible solution set is a collection of autonomy levels defined according to different ranges of autonomy capabilities; that is, the feasible solution set is... Considering factors such as the difficulty of completing the task and the degree of achievement of the task objectives, four preliminary plans have been proposed for the collaborative detection phase, as shown below: Option 1 In this scheme, the operator's role shifts from direct operator to command issuer, responsible for issuing high-level maneuvering and detection mode commands, and performing situational assessments based on system-processed information. The auxiliary system begins to generate specific control data and can perform preliminary classification and anomaly labeling of sensor data. The aircraft is able to execute more abstract commands and possesses basic autonomous obstacle avoidance and sensor coordination capabilities. The interaction relationship transforms into a "command-execution" mode, and the system begins to provide preliminary automated assistance.

[0089] Option 2 Under this scheme, the operator's role is elevated to the mission management level, primarily responsible for setting planned routes and collaborative detection modes, and approving various path optimization schemes generated by the auxiliary system. The auxiliary system can automatically generate optimization schemes based on environmental threats and initially integrate multi-sensor data to mark high-value targets. The aircraft can autonomously execute the approved path and optimize the allocation of sensor resources. The core interaction at this level is "human on the loop," meaning the operator reviews and makes decisions regarding system recommendations.

[0090] Option 3 Under this scheme, the operator primarily sets mission-level objectives for flight and reconnaissance, and monitors the overall system operation and data fusion quality, intervening only in cases of anomalies. The auxiliary system is fully authorized to autonomously plan flight paths, adjust formations, allocate sensor cooperative detection modes, and generate preliminary threat assessments. The aircraft can autonomously execute formation cooperative maneuvers and synchronize sensor data between aircraft. The interaction relationship evolves into a "supervision-delegation" model, with the system operating highly autonomously within its authorized scope.

[0091] Option 4 Under this scheme, operators only need to set strategic objectives and intelligence requirements, and monitor the overall reliability and data trustworthiness of the system; the auxiliary system, based on AI models, can autonomously plan dynamic task chains, intelligently identify targets, and dynamically reconstruct the sensor network; the aircraft possesses advanced swarm intelligence, enabling it to autonomously adapt to complex environments and initiate coordinated actions and sensor task strategies. This level of interaction manifests as "human-outside-the-loop" strategic supervision, with the system dominating the vast majority of decisions and execution.

[0092] Step 2: Establish the decision matrix and determine the weight coefficients. The human-machine permission allocation level decision-making process utilizes multi-objective attribute functions to obtain indicator attribute values ​​under different permission allocation levels. Then, a decision-making algorithm is used to calculate the human-machine permission allocation level: After constructing a multi-dimensional evaluation index system, the Analytic Hierarchy Process (AHP) is used to allocate index weights. This process includes two core stages: First, single-level index weights are calculated by introducing a 1-9 scale to compare pairs of indices belonging to the same criterion, constructing a pairwise comparison judgment matrix; then, the eigenvalue method is used to solve for the largest eigenvalue and its corresponding eigenvector of this matrix, completing the weight vector extraction. Special emphasis should be placed on the matrix consistency check stage, verifying the self-consistency of the judgment logic by calculating the consistency ratio (CR). If the decision matrix exhibits acceptable consistency, then a revised decision matrix must be returned. The mathematical expression for constructing the pairwise comparison matrix between indicators is shown in the following formula. This process strictly follows the AHP theoretical framework to ensure the objectivity and systematic nature of the weight allocation.

[0093]

[0094] After obtaining the judgment matrix, its largest eigenvalue is determined by solving the characteristic equation. And extract the corresponding feature vectors. .

[0095] After normalizing the feature vector, the single-layer index weight vector can be obtained. Based on the Random Consistency Index (RI) reference system provided in Table 1.4, the consistency index is calculated using the formula. Furthermore, the consistency ratio was derived through a formula. Verification has shown that the criterion-level judgment matrix constructed in this invention strictly satisfies... The decision reliability requirement indicates that the values ​​of matrix elements are logically consistent and that the weight configuration results have effective decision value.

[0096] (1) Task completion difficulty judgment matrix The judgment matrix of the difficulty of human / machine collaborative task completion relative to the target layer is shown in the following formula:

[0097] After obtaining the judgment matrix, its largest eigenvalue is determined by solving the characteristic equation. And extract the corresponding feature vectors. After normalizing the feature vector, the single-layer index weight vector can be obtained. Based on the Random Consistency Index (RI) reference system provided in Table 1.4, the consistency index is calculated using the formula. Furthermore, the consistency ratio was derived through a formula. Verification has shown that the criterion-level judgment matrix constructed in this study strictly satisfies... The decision reliability requirement indicates that the values ​​of matrix elements are logically consistent and that the weight configuration results have effective decision value.

[0098] (2) Task objective completion judgment matrix The judgment matrix of the human / machine collaborative task objective completion degree relative to the objective layer is shown in the following formula:

[0099] After obtaining the judgment matrix, its largest eigenvalue is determined by solving the characteristic equation. And extract the corresponding feature vectors. After normalizing the feature vector, the single-layer index weight vector can be obtained. Based on the Random Consistency Index (RI) reference system provided in Table 1.4, the consistency index is calculated using the formula. Furthermore, the consistency ratio was derived through a formula. Verification has shown that the criterion-level judgment matrix constructed in this study strictly satisfies... The decision reliability requirement indicates that the values ​​of matrix elements are logically consistent and that the weight configuration results have effective decision value.

[0100] (3) Environmental threat level assessment matrix The judgment matrix of the threat level of the human / machine collaborative task environment relative to the target layer is shown in the following formula:

[0101] After obtaining the judgment matrix, its largest eigenvalue is determined by solving the characteristic equation. And extract the corresponding feature vectors. After normalizing the feature vector, the single-layer index weight vector can be obtained. Based on the Random Consistency Index (RI) reference system provided in Table 1.4, the consistency index is calculated using the formula. Furthermore, the consistency ratio was derived through a formula. Verification has shown that the criterion-level judgment matrix constructed in this study strictly satisfies... The decision reliability requirement indicates that the values ​​of matrix elements are logically consistent and that the weight configuration results have effective decision value.

[0102] (4) Spatial Threat Assessment Matrix The judgment matrix of the spatial threat level of human / machine collaborative tasks relative to the target layer is shown in the following formula:

[0103] After obtaining the judgment matrix, its largest eigenvalue is determined by solving the characteristic equation. And extract the corresponding feature vectors. After normalizing the feature vector, the single-layer index weight vector can be obtained. Based on the Random Consistency Index (RI) reference system provided in Table 1.4, the consistency index is calculated using the formula. Furthermore, the consistency ratio was derived through a formula. Verification has shown that the criterion-level judgment matrix constructed in this embodiment strictly satisfies... The decision reliability requirement indicates that the values ​​of matrix elements are logically consistent and that the weight configuration results have effective decision value.

[0104] (5) Detection capability judgment matrix The judgment matrix of human / machine collaborative task detection capability relative to the target layer is shown in the following formula:

[0105] After obtaining the judgment matrix, its largest eigenvalue is determined by solving the characteristic equation. And extract the corresponding feature vectors. After normalizing the feature vector, the single-layer index weight vector can be obtained. Based on the Random Consistency Index (RI) reference system provided in Table 1.4, the consistency index is calculated using the formula. Furthermore, the consistency ratio was derived through a formula. Verification has shown that the criterion-level judgment matrix constructed in this study strictly satisfies... The decision reliability requirement indicates that the values ​​of matrix elements are logically consistent and that the weight configuration results have effective decision value.

[0106] (6) Anti-interference capability judgment matrix The judgment matrix of the anti-interference capability of human / machine collaborative tasks relative to the target layer is shown in the following formula:

[0107] After obtaining the judgment matrix, its largest eigenvalue is determined by solving the characteristic equation. And extract the corresponding feature vectors. After normalizing the feature vector, the single-layer index weight vector can be obtained. Based on the Random Consistency Index (RI) reference system provided in Table 1.4, the consistency index is calculated using the formula. Furthermore, the consistency ratio was derived through a formula. Verification has shown that the criterion-level judgment matrix constructed in this study strictly satisfies... The decision reliability requirement indicates that the values ​​of matrix elements are logically consistent and that the weight configuration results have effective decision value.

[0108] (7) Command and decision-making ability judgment matrix The judgment matrix of human / machine collaborative task command and decision-making ability relative to the target layer is shown in the following formula:

[0109] After obtaining the judgment matrix, its largest eigenvalue is determined by solving the characteristic equation. And extract the corresponding feature vectors. After normalizing the feature vector, the single-layer index weight vector can be obtained. Based on the Random Consistency Index (RI) reference system provided in Table 1.4, the consistency index is calculated using the formula. Furthermore, the consistency ratio was derived through a formula. Verification has shown that the criterion-level judgment matrix constructed in this study strictly satisfies... The decision reliability requirement indicates that the values ​​of matrix elements are logically consistent and that the weight configuration results have effective decision value.

[0110] (8) Decision-making ability assessment matrix The judgment matrix of human / machine collaborative task auxiliary decision-making ability relative to the target layer is shown in the following formula:

[0111] After obtaining the judgment matrix, its largest eigenvalue is determined by solving the characteristic equation. And extract the corresponding feature vectors. After normalizing the feature vector, the single-layer index weight vector can be obtained. Based on the Random Consistency Index (RI) reference system provided in Table 1.4, the consistency index is calculated using the formula. Furthermore, the consistency ratio was derived through a formula. Verification has shown that the criterion-level judgment matrix constructed in this study strictly satisfies... The decision reliability requirement indicates that the values ​​of matrix elements are logically consistent and that the weight configuration results have effective decision value.

[0112] Finally, construct the final weight table and fill in the results obtained by the eigenvalue method into the table, as shown in Table 2.1.

[0113] Table 2.1 Weighting of Indicators for Human / Machine Collaborative Systems

[0114] After determining the weights of the individual indicators, the total weights are calculated from the individual indicator weights. The calculation results are shown in Table 2.2.

[0115] Table 2.2 Total Weight Table of Human / Machine Collaborative System

[0116] Step 3: Solve for permission allocation levels Solving for permission allocation levels using a comprehensive weighted method: (1) Each decision-maker makes a multi-indicator decision on each option individually, assuming the first option is the first option. The aircraft to the first The first scheme The target evaluation value is The evaluation values ​​were determined using a standardized scoring system, resulting in an evaluation matrix. In the calculation process, this involves assigning priority levels to individual entities within a multi-aircraft fleet based on their current priority authority. .

[0117] (2) Calculate the arithmetic weighted average score and the geometric weighted average score for each scheme. The arithmetic weighted average score is... , For the first The weight values ​​of each objective. The geometrically weighted average is divided into... .

[0118] (3) Calculate the mixed average score for each scheme. .

[0119] (4) Calculate the overall average score for each scheme. .

[0120] (5) Calculate the evaluation coefficients for each scheme. The evaluation values ​​of the indicators for each scheme in the search and engagement phase are shown in Table 2.3, calculated according to the formula.

[0121] Table 2.3 Evaluation values ​​of each scheme's evaluation indicators

[0122] (6) Based on the evaluation coefficient Arrange the priority levels of the permission allocation schemes from highest to lowest. Based on the total weight obtained and the evaluation values ​​of each scheme's evaluation indicators, the comprehensive score of each scheme is calculated, as shown in Table 2.4.

[0123] Table 2.4 Overall Score Table for Each Option

[0124] Based on the above calculations, during the collaborative detection phase, the human / machine permission allocation level under Scheme 4 has the highest overall score, while the permission allocation level under Scheme 2 has the lowest overall score.

[0125] Any part of this invention not described in detail can be referred to in the prior art or in the art known to those skilled in the art. This embodiment does not limit such part and will not describe it in detail here.

[0126] The embodiments of the present invention have been described above with reference to the accompanying drawings. However, the present invention is not limited to the specific embodiments described above. The specific embodiments described above are merely illustrative and not restrictive. Those skilled in the art can make many other forms under the guidance of the present invention without departing from the spirit and scope of the claims, and all of these forms are within the protection scope of the present invention.

Claims

1. A method for allocating permissions for collaborative missions of aircraft based on multi-attribute decision-making, characterized in that, The method includes: S1. Hierarchical Design of Auxiliary Systems: Divide the autonomous levels into tiered levels, clarify the responsibilities of operators, auxiliary systems and aircraft at different levels, define each level as a solution, and form a solution set; S2. Target attribute modeling: Transform the key factors affecting permission allocation into calculable quantitative indicators to avoid subjective decision-making; S3. Establish a decision matrix: Transform the quantitative results into a correspondence between permission level and attribute cognitive quality to obtain a decision matrix, providing a data carrier for subsequent decision-making; S4. Determine the weighting coefficients: Determine the evaluation indicators and assign reasonable weights to each evaluation indicator to reflect the differences in the importance of the indicators; S5. Solve for permission allocation levels: Based on the decision matrix and weights, calculate the comprehensive score of each permission level, sort them, and determine the optimal level.

2. The method for allocating permissions for collaborative aircraft tasks based on multi-attribute decision-making according to claim 1, characterized in that, The levels in step S1 include auxiliary control level, semi-autonomous level, highly autonomous level and fully autonomous level.

3. The method for allocating permissions for collaborative aircraft tasks based on multi-attribute decision-making according to claim 2, characterized in that, The operator's responsibilities in the auxiliary control level are: issuing flight maneuver and detection mode commands, and assessing the overall situation; the auxiliary system functions are: generating control data, monitoring flight status, and marking sensor anomalies. The aircraft's capabilities include: executing commands and basic obstacle avoidance; In the semi-autonomous stage, the operator's responsibilities are: setting the route and detection mode, and selecting the path optimization scheme; the auxiliary system functions are: automatically generating control data, optimizing the formation path, and marking high-value targets; the aircraft's capabilities are: executing the approved path and optimizing sensor allocation; The operator's responsibilities in the highly autonomous level are: autonomously planning paths and adjusting formations, allocating sensor modes, and generating threat assessments; the auxiliary system functions are: autonomous coordinated actions and inter-aircraft data synchronization; the aircraft's capabilities are: autonomous coordinated actions and inter-aircraft data synchronization. In the fully autonomous class, the operator's responsibilities are: setting mission objectives and intelligence requirements, and monitoring the credibility of data; the auxiliary system functions are: planning dynamic mission chains, target identification, and reconstructing sensor networks; and the aircraft's capabilities are: swarm intelligence and autonomous adaptation to complex environments.

4. The method for allocating permissions for collaborative aircraft tasks based on multi-attribute decision-making according to claim 1, characterized in that, Step S2 includes: S21. Multi-attribute decision problem modeling: Solve the priority ranking or evaluation problem of several feasible solutions under multiple objectives by solving the complete order of feasible solutions; S22. Task Attribute Modeling: Decompose task attribute information into task objective difficulty and task execution status; establish a task objective difficulty evaluation model for task objective difficulty and a task objective achievement evaluation model for task execution status to evaluate the current task execution completion level; S23. Environmental Attribute Modeling: Divide the left and right sides of the aircraft into fan-shaped regions and construct a threat level model to quantify the threat level of each fan-shaped region to the aircraft; S24. Aircraft capability modeling: Quantify the aircraft's detection capabilities and anti-jamming capabilities separately; S25. Operator Command Decision Attribute Modeling: Quantitatively model command decision preferences using survival priority coefficients, and introduce decision load coefficients as performance evaluation indicators to model auxiliary decision-making capabilities.

5. The method for allocating permissions for collaborative aircraft tasks based on multi-attribute decision-making according to claim 4, characterized in that, The detection capability of aircraft in S24 is quantified by coverage factor, overlap factor and jamming suppression ratio. The anti-jamming capability is quantified using antenna gain, radar average power, signal bandwidth, signal illumination time, target reflective area, and radar resolution volume unit.

6. The method for allocating permissions for collaborative aircraft tasks based on multi-attribute decision-making according to claim 1, characterized in that, Step S3 includes: Matrix element definition: Let x be the cognitive quality of the i-th attribute at the j-th permission level. ij Cognitive quality x ij This indicates the aircraft's level of awareness of attribute i at level j; Introducing a utility function: linking the probability of user intervention with the quality of attribute cognition, and ensuring that the quality of cognition can be quantified into decision utility through a pre-defined formula.

7. The method for allocating permissions for collaborative aircraft tasks based on multi-attribute decision-making according to claim 1, characterized in that, Step S4 includes: S41. Provide a detailed description of the difficulty of mission completion, the degree of mission objective completion, the degree of environmental threat, the degree of space threat, detection capability, anti-jamming capability, command and decision-making capability, and auxiliary decision-making capability, and obtain evaluation values ​​for the environment, mission, user, and aircraft capabilities; S42. Multiple performance evaluation indicators: Evaluation functions are used to evaluate each indicator to obtain a decision matrix; S43. Calculate the optimal permission allocation level based on the human-machine permission allocation level adjustment algorithm.

8. The method for allocating permissions for collaborative aircraft tasks based on multi-attribute decision-making according to claim 7, characterized in that, Step S43 uses the comprehensive weighted method in group decision-making to solve for the level of authority allocation.

9. The method for allocating permissions for collaborative aircraft tasks based on multi-attribute decision-making according to claim 8, characterized in that, Step S43 includes the following steps: Determine the weights of different decision attributes; Calculate the priority value for each pair of permission assignment levels; The priority order of the current permission allocation level is obtained based on the priority value; The evaluation matrix is ​​composed of the priority order calculated from multiple aircraft in the cluster.

10. A computer device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that the processor executes the computer program to implement the steps of the aircraft cooperative mission authority allocation method based on multi-attribute decision-making as described in any one of claims 1-9.

Images