Search

CN-121256715-B - Intelligent generation method and device for unmanned aerial vehicle countering strategy

CN121256715BCN 121256715 BCN121256715 BCN 121256715BCN-121256715-B

Abstract

The invention discloses an intelligent generation method and device of an unmanned aerial vehicle countering strategy, relates to the technical field of unmanned aerial vehicles, and aims to solve the problems of poor countering means matching effect and low efficiency caused by limited countering applicable scenes of unmanned aerial vehicles, lack of deep semantic support and single means in the prior art. The method comprises the steps of obtaining multi-source sensor data of a target area, determining environment information of the target area and situation type information of a suspicious unmanned aerial vehicle, extracting features to be matched corresponding to the situation type information from the multi-source sensor data, performing similarity matching with features of a plurality of target matching entity nodes respectively, determining a matching result, obtaining at least one countermeasures from a countermeasures knowledge graph of the unmanned aerial vehicle and generating an initial countermeasures strategy set based on the matching result and the environment information, and processing the initial countermeasures strategy set by adopting a target optimization algorithm to obtain a target countermeasures strategy. The invention is used for improving the matching precision and the matching efficiency of the unmanned aerial vehicle countering technology.

Inventors

  • LIU KUILIN

Assignees

  • 融鼎岳(北京)科技有限公司

Dates

Publication Date
20260505
Application Date
20251204

Claims (9)

  1. 1. The intelligent generation method of the unmanned aerial vehicle countering strategy is characterized by comprising the following steps of: Acquiring multi-source sensor data of a target area, and determining environmental information of the target area and situation type information of a suspicious unmanned aerial vehicle based on the multi-source sensor data, wherein the situation type information comprises a single machine situation and a cluster situation; Extracting characteristics to be matched corresponding to the situation type information from the multi-source sensor data, and respectively carrying out similarity matching on the characteristics to be matched and characteristics of a plurality of target matching entity nodes in an unmanned aerial vehicle countercheck knowledge graph to determine a matching result, wherein the characteristics to be matched are multi-modal characteristics to be matched corresponding to the single-machine situation or cluster situation characteristics to be matched corresponding to the cluster situation; Based on the matching result and the environmental information, acquiring at least one countermeasures from the unmanned aerial vehicle countermeasures knowledge graph and generating an initial countermeasures strategy set; processing the initial countering strategy set by adopting a target optimization algorithm to obtain a target countering strategy; The entity nodes in the unmanned aerial vehicle reaction knowledge graph at least comprise tactical model nodes; when the situation type information is a cluster situation, the target matching entity node is a tactical model node, and the matching result is a target tactical model node; The cluster situation features to be matched comprise a quantity structure feature, a formation feature, a cooperative action feature and a communication cooperative feature; The tactical mode node at least comprises a cluster cooperative attack mode node, a multi-type mixed detour anti-burst mode node and a distributed cluster harassment mode node; Determining a matching result based on similarity values corresponding to all target matching entity nodes, including: and determining the target tactical mode node by using tactical mode nodes corresponding to maximum similarity values.
  2. 2. The intelligent generation method of unmanned aerial vehicle countering strategies according to claim 1, wherein, The features to be matched are respectively subjected to similarity matching with the features of a plurality of target matching entity nodes in the unmanned aerial vehicle reaction knowledge graph, and a matching result is determined, wherein the steps include: According to a similarity calculation formula: ; Calculating the similarity value corresponding to each target matching entity node, wherein, Is the similarity; The feature vector is to be matched; ; matching the feature vector of the entity node for the target; ; Is the first of the feature vectors to be matched Sub-feature items; Matching the first of the feature vectors of the entity node for the target Sub-feature items; Is the first The weight coefficient of the sub-feature item; Is a positive integer greater than 2; and determining a matching result based on the similarity values corresponding to all the target matching entity nodes.
  3. 3. The intelligent generation method of the unmanned aerial vehicle countering strategy according to claim 2, wherein the entity nodes in the unmanned aerial vehicle countering knowledge graph further comprise unmanned aerial vehicle model nodes, when the situation type information is the single-machine situation, the target matching entity nodes are unmanned aerial vehicle model nodes, and the matching result is target unmanned aerial vehicle model nodes matched with the suspicious unmanned aerial vehicle; Determining a matching result based on similarity values corresponding to all target matching entity nodes, including: Comparing the similarity degree values corresponding to all the unmanned aerial vehicle model nodes with a first preset similarity threshold value respectively; If the similarity value corresponding to at least one unmanned aerial vehicle model node is larger than or equal to the first preset similarity threshold value, determining the unmanned aerial vehicle model node corresponding to the maximum value of the similarity values as the target unmanned aerial vehicle model node; and if the similarity values corresponding to all the unmanned aerial vehicle model nodes are smaller than the first preset similarity threshold, judging that the model of the suspicious unmanned aerial vehicle is an unknown model, and continuing to match according to a preset priority order to determine the target unmanned aerial vehicle model node, wherein the preset priority order is a communication protocol node, a sub-communication protocol node, a hardware feature class node and a software feature class node in sequence.
  4. 4. The intelligent generation method of unmanned aerial vehicle countering strategies according to claim 3, wherein the entity nodes in the unmanned aerial vehicle countering knowledge graph further comprise a plurality of communication protocol nodes; Continuing to match according to a preset priority order to determine the target unmanned aerial vehicle model node, wherein the method comprises the following steps of: Extracting a plurality of communication protocol sub-features in the multi-mode features to be matched and constructing a communication protocol feature vector, wherein the plurality of communication protocol sub-features comprise signal frequency bands, frequency hopping rules and frame structure identifiers; adopting the similarity calculation formula, respectively carrying out similarity matching on the communication protocol feature vector and preset feature vectors of a plurality of communication protocol nodes, and calculating to obtain a first similarity result of each communication protocol node; if the first similarity result of at least one communication protocol node is larger than or equal to a second preset similarity threshold value, determining the communication protocol node corresponding to the maximum value of the first similarity result as a target communication protocol node, and determining the node of the known unmanned aerial vehicle model corresponding to the target communication protocol node as the target unmanned aerial vehicle model node through the adoption of a protocol relationship between the communication protocol node and the unmanned aerial vehicle model node; If the first similarity results of all the communication protocol nodes are smaller than the second preset similarity threshold, invoking a similar protocol relation among the communication protocol nodes in the unmanned aerial vehicle reaction knowledge graph, and screening a plurality of sub-communication protocol nodes associated with the target communication protocol node based on a screening condition to continuously perform similarity matching to obtain a plurality of second similarity results, wherein the screening condition is that a protocol similarity attribute value is larger than or equal to a third preset similarity threshold; If at least one second similarity result is larger than or equal to the second preset similarity threshold value, obtaining a corresponding target unmanned aerial vehicle model node; And if all the second similarity results are smaller than the second preset similarity threshold, continuing to match according to the preset priority order to determine the target unmanned aerial vehicle model node.
  5. 5. The intelligent generation method of unmanned aerial vehicle countering strategies according to claim 2, wherein the relationships in the unmanned aerial vehicle countering knowledge graph at least comprise a countering relationship, an applicable environment relationship, a forbidden environment relationship and an environment influence countering validity relationship; when the situation type information is a single-machine situation, based on the matching result and the environmental information, obtaining at least one countermeasures from the unmanned aerial vehicle countermeasures knowledge graph and generating an initial countermeasures strategy set, wherein the method comprises the following steps: acquiring all countermeasures associated with the matching result through the countermeasures to form a first countermeasure pool; Screening out the reaction means containing the environmental information from the first reaction means pool through the applicable environmental relation to obtain a second reaction means pool; Removing the reaction means of which the forbidden environment contains the environment information in the second reaction means pool through the forbidden environment relation to obtain a third reaction means pool; According to the environment influence counteraction effectiveness relation, the association attribute of the environment information in the unmanned aerial vehicle counteraction knowledge graph and each counteraction means in the third counteraction means pool is called, and the corrected effectiveness is calculated by combining the original effectiveness of each counteraction means; based on the corrected effectiveness of each means in the third reactive means pool, combining all the reactive means and marking the expected reactive effect of each combination to obtain an initial reactive strategy set of a single machine situation.
  6. 6. The intelligent generation method of unmanned aerial vehicle countering strategies according to claim 5, wherein the relationships in the unmanned aerial vehicle countering knowledge graph further comprise tactical combination relationships; When the situation type information is a cluster situation, based on the matching result and the environmental information, acquiring at least one countermeasures from the unmanned aerial vehicle countermeasures knowledge graph and generating an initial countermeasures strategy set, wherein the method comprises the following steps: invoking tactical combination relations to obtain a countermeasures collaborative framework associated with the target tactical mode node; Aiming at each link of the countermeasures collaborative framework, acquiring candidate countermeasures of each link from the unmanned aerial vehicle countermeasures knowledge graph; Screening means including the environment information in the applicable environments in the candidate countermeasures of each link based on the applicable environment relation to obtain a first countermeasure set of each link; Based on the forbidden environment relation, excluding the forbidden environment in the first countermeasures set of each link from the measures containing the environment information, and obtaining a second countermeasures set of each link; Based on the environment influence reaction effectiveness relation, the association attribute of each means in the second reaction means set of each link is called, the original effectiveness of each means is corrected, and the correction effectiveness of each means is obtained; And combining correction effectiveness of each means, selecting means from a second countermeasures set of each link, combining the means according to the time sequence of the countermeasures and the frame, marking the expected countermeasures of each combination, and generating an initial countermeasures strategy set of the cluster situation.
  7. 7. The intelligent generation method of the unmanned aerial vehicle countering strategy according to claim 1, wherein the processing the initial countering strategy set by using a target optimization algorithm to obtain a target countering strategy comprises: Determining a multi-objective optimization set, wherein the multi-objective optimization set comprises benefit-type optimization targets and cost-type optimization targets, the benefit-type optimization targets comprise reverse power and collaborative efficiency, and the cost-type optimization targets comprise response time, economic cost and collateral damage; the formula is adopted: ; Normalizing the original data of the benefit type optimization targets to obtain normalized values of the benefit type optimization targets, wherein, Is the first Normalized values of individual benefit optimization objectives; Is the first Raw data of the individual benefit optimization objective; Is the first Minimum reference value of individual benefit optimization target; Is the first Maximum reference value of individual benefit optimization target; the formula is adopted: ; Normalizing the original data of the cost type optimization targets to obtain normalized values of the cost type optimization targets, wherein, Is the first Normalized values of the individual cost-effective optimization objectives; Is the first Raw data of the individual cost-effective optimization objectives; Is the first Minimum reference value of each cost-type optimization target; Is the first Maximum reference value of each cost-type optimization target; Determining weight coefficients of all optimization targets according to the situation type information; the formula is adopted: ; A composite score is calculated for each initial countering strategy in the set of initial countering strategies, wherein, A score for the composite; Is the first Weight coefficient of each optimization target, if The optimization targets are benefit-type optimization targets And, if you get The optimization targets are cost-type optimization targets ; And selecting the initial countering strategy with the highest comprehensive score as a target countering strategy.
  8. 8. The intelligent generation method of unmanned aerial vehicle countering strategies according to claim 1, wherein, Further comprises: Controlling corresponding countering executing equipment to execute a countering task on the suspicious unmanned aerial vehicle based on the target countering strategy, and recording countering process data in the executing process of the countering task in real time; And if the countering process data show that the countering effectiveness of the target countering strategy is higher than a preset threshold, updating the attribute of the countering effectiveness relation between the target countering strategy and the corresponding unmanned aerial vehicle model number in the unmanned aerial vehicle countering knowledge graph.
  9. 9. Intelligent generation device of unmanned aerial vehicle countering strategy, its characterized in that includes: the system comprises an acquisition module, a storage module and a processing module, wherein the acquisition module is used for acquiring multi-source sensor data of a target area and determining environmental information of the target area and situation type information of suspicious unmanned aerial vehicles based on the multi-source sensor data, wherein the situation type information comprises a single-machine situation and a cluster situation; The matching module is used for extracting characteristics to be matched corresponding to the situation type information from the multi-source sensor data, and respectively carrying out similarity matching on the characteristics to be matched and the characteristics of a plurality of target matching entity nodes in the unmanned aerial vehicle countercheck knowledge graph to determine a matching result; the generation module is used for acquiring at least one countermeasures from the unmanned aerial vehicle countermeasures knowledge graph and generating an initial countermeasures strategy set based on the matching result and the environmental information; The optimization module is used for processing the initial countering strategy set by adopting a target optimization algorithm to obtain a target countering strategy; The entity nodes in the unmanned aerial vehicle reaction knowledge graph at least comprise tactical model nodes; when the situation type information is a cluster situation, the target matching entity node is a tactical model node, and the matching result is a target tactical model node; The cluster situation features to be matched comprise a quantity structure feature, a formation feature, a cooperative action feature and a communication cooperative feature; The tactical mode node at least comprises a cluster cooperative attack mode node, a multi-type mixed detour anti-burst mode node and a distributed cluster harassment mode node; Determining a matching result based on similarity values corresponding to all target matching entity nodes, including: and determining the target tactical mode node by using tactical mode nodes corresponding to maximum similarity values.

Description

Intelligent generation method and device for unmanned aerial vehicle countering strategy Technical Field The invention relates to the technical field of unmanned aerial vehicles, in particular to an intelligent generation method and device of an unmanned aerial vehicle countering strategy. Background With the rapid development of unmanned aerial vehicle technology, security threats such as illegal invasion, malicious reconnaissance and the like are increasingly prominent, and an unmanned aerial vehicle countering system becomes key equipment for guaranteeing airspace security. The existing unmanned aerial vehicle countering strategy generation method mainly depends on a preset rule base or simple numerical matching, and has obvious defects in practical application: First, the adaptability to complex scenes is poor. The existing method is difficult to associate various factors such as unmanned aerial vehicle situation, flight environment and the like, and when facing an unknown model unmanned aerial vehicle or a novel tactic, an effective countercheck strategy cannot be generated based on characteristic association, only preset rules can be passively matched, and the adaptability is limited. Secondly, decisions lack deep semantic support. The existing method only determines a countermeasures through numerical calculation or surface feature matching, and cannot mine the associated logic behind the data, so that the countermeasures of the countermeasures are insufficient in pertinence of the countermeasures, and the diversified unmanned aerial vehicle threats are difficult to deal with. Third, the reaction scheme is singulated. The existing system can only output a single countering means generally, cannot form a systematic scheme matched with multiple means, is difficult to balance countering effect and scene safety, and is easy to cause secondary risks in complex environments such as airports, cities and the like. Therefore, there is a need for an unmanned aerial vehicle countering strategy generation method that can adapt to complex scenarios, mine deep association, and generate multiple schemes, so as to solve the above-mentioned drawbacks of the prior art. Disclosure of Invention The invention aims to provide an intelligent generation method and device of an unmanned aerial vehicle countering strategy, which are used for improving the countering accuracy and the countering efficiency of an unmanned aerial vehicle. In order to achieve the above object, the present invention provides the following technical solutions: In a first aspect, the present invention provides an intelligent generation method for a countering strategy of an unmanned aerial vehicle, including: Acquiring multi-source sensor data of a target area, and determining environmental information of the target area and situation type information of a suspicious unmanned aerial vehicle based on the multi-source sensor data, wherein the situation type information comprises a single machine situation and a cluster situation; Extracting characteristics to be matched corresponding to the situation type information from the multi-source sensor data, and respectively carrying out similarity matching on the characteristics to be matched and characteristics of a plurality of target matching entity nodes in an unmanned aerial vehicle countercheck knowledge graph to determine a matching result, wherein the characteristics to be matched are multi-modal characteristics to be matched corresponding to the single-machine situation or cluster situation characteristics to be matched corresponding to the cluster situation; Based on the matching result and the environmental information, acquiring at least one countermeasures from the unmanned aerial vehicle countermeasures knowledge graph and generating an initial countermeasures strategy set; and processing the initial countering strategy set by adopting a target optimization algorithm to obtain a target countering strategy. Optionally, performing similarity matching on the features to be matched and features of a plurality of target matching entity nodes in the unmanned aerial vehicle reaction knowledge graph, and determining a matching result includes: According to a similarity calculation formula: ; Calculating the similarity value corresponding to each target matching entity node, wherein, Is the similarity; The feature vector is to be matched; ; matching the feature vector of the entity node for the target; ; Is the first of the feature vectors to be matched Sub-feature items; Matching the first of the feature vectors of the entity node for the target Sub-feature items; Is a positive integer greater than 2; Is the first The weight coefficient of the sub-feature item; Is a positive integer greater than 2; and determining a matching result based on the similarity values corresponding to all the target matching entity nodes. Optionally, the entity nodes in the unmanned aerial vehicle countering knowledge graph