Airplane intent identification probability estimation method and system

Abstract

The application provides an aircraft intention recognition probability estimation method and system, and relates to the technical field of aircraft intention recognition. Basic information of a target aircraft is acquired, attribute information and state information are determined in the basic information; semantic information fuzzy sets are generated according to the attribute information and the state information to serve as antecedent structure parameters; target task information is acquired, and membership functions in consequent structures are determined according to the target task information; the intention probability prediction model of the target aircraft is generated by combining the basic information and the target task information and through aggregation of preset semantic rules; the basic information of the current target aircraft is acquired, and the flight intention is determined in combination with the intention probability prediction model. Through multiple rules, complex nonlinear relationships are approximated, various complex data are fitted, each rule has semantic characteristics, and strong interpretability can be achieved, so that the technical effect of efficiently and accurately identifying the aircraft intention is realized.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of aircraft intention recognition, in particular to an aircraft intention recognition probability estimation method and system. BACKGROUND

[0002] Target intention refers to a certain purpose that a target unit hopes to achieve or a state that the target unit expects to realize, and it is highly related to the actual actions and state of the target unit. Different aircraft target trajectories usually represent different intentions. By monitoring aircraft elements and attributes and performing feature extraction and analysis to obtain the change rule of data, target intention prediction can be achieved.

[0003] The current commonly used method of intention recognition can be divided into two categories. One is to realize reasoning based on prior knowledge. Traditional intention prediction methods mainly rely on the knowledge base of prior knowledge and logical reasoning ability to realize intention judgment. Prior knowledge includes domain expert knowledge, historical data and general natural laws. Domain experts form a deep understanding of a specific field through long-term practice and experience accumulation; historical data contains action rules and potential information, which is beneficial to infer future trends; natural laws as objective facts can guide and assist the reasoning process. This method has strong reliability and interpretability, but poor flexibility. At the same time, different expert knowledge is highly subjective, and it is very complex to convert it into a computable and understandable form, and there are subjective bias problems. The other is a machine learning method without prior conditions. Neural network-based methods are widely used in intention recognition. This kind of method does not need to define rules or knowledge in advance, and can realize aircraft target feature extraction and information mining from data by using machine learning algorithm, and automatically learn the mode and rule of aircraft target in complex environment to infer the intention of aircraft target. At the same time, this method can continuously improve the performance with the accumulation of effective data. However, in the process of intention judgment, the internal decision-making process of neural network method is complex, and the whole process is difficult to explain, and the credibility and reliability of the model are low. In addition, the training and deployment of complex machine models require high computing resources and time, and the feasibility in practical application is limited. In summary, although the existing two methods perform well in some aspects, they all have some limitations. The method based on prior knowledge has large data demand, strong subjective factors and high practical application requirements; the machine learning method is difficult to explain and has poor robustness.

[0004] Therefore, how to efficiently and accurately recognize the intention of the aircraft has become a technical problem to be solved. SUMMARY

[0005] In order to efficiently and accurately recognize the intention of the aircraft, the present application provides an aircraft intention recognition probability estimation method and system.

[0006] In a first aspect, the application provides an aircraft intention recognition probability estimation method using the following technical solution:

[0007] An aircraft intention recognition probability estimation method comprises:

[0008] Obtaining basic information of a target aircraft, determining attribute information and state information in the basic information;

[0009] Generating a semantic information fuzzy set as an antecedent structure parameter according to the attribute information and the state information;

[0010] Obtaining target task information and determining a membership function in a consequent structure according to the target task information;

[0011] Combining the basic information and the target task information, generating an intention probability prediction model of the target aircraft by aggregating preset semantic rules;

[0012] Obtaining current basic information of the target aircraft and determining a flight intention in combination with the intention probability prediction model.

[0013] Optionally, the step of generating a semantic information fuzzy set as an antecedent structure parameter according to the attribute information and the state information comprises:

[0014] Generating an attribute set according to the attribute information;

[0015] Generating a state set according to the state information;

[0016] Generating a semantic information fuzzy set as an antecedent structure parameter in combination with the attribute set and the state set according to a semantic information screening strategy.

[0017] Optionally, the step of obtaining target task information and determining a membership function in a consequent structure according to the target task information comprises:

[0018] Obtaining target task information and determining judgment conclusion information according to the target task information;

[0019] Determining whether the number of conclusions in the judgment conclusion information satisfies a preset conclusion condition;

[0020] If yes, determining a membership function in a consequent structure according to the judgment conclusion information.

[0021] Optionally, the step of determining a membership function in a consequent structure according to the judgment conclusion information comprises:

[0022] Determining a conclusion probability set in the judgment conclusion information;

[0023] determining a target parameter type in the semantic information fuzzy set according to the conclusion probability set in a semantic reasoning manner;

[0024] establishing a membership function in a consequent structure according to the target parameter type.

[0025] Optionally, the step of establishing the membership function in the consequent structure according to the target parameter type comprises:

[0026] establishing a rule set of the semantic information fuzzy set according to the target parameter type;

[0027] performing rationality screening in the rule set to generate a target rule set;

[0028] generating a membership function in a consequent structure by combining the target rule set in a rule aggregation manner.

[0029] Optionally, the step of acquiring the basic information of the target aircraft and determining the flight intention in combination with the intention probability prediction model comprises:

[0030] acquiring the basic information of the target aircraft to generate an input parameter set;

[0031] inputting the input parameter set into the intention probability prediction model and acquiring an output result;

[0032] acquiring a conclusion and corresponding probability information in the output result;

[0033] determining the flight intention according to the conclusion and the corresponding probability information.

[0034] Optionally, the step of determining the flight intention according to the conclusion and the corresponding probability information comprises:

[0035] acquiring a preset determination threshold and determining whether there is probability information satisfying the preset determination threshold in the output result;

[0036] if yes, determining the flight intention according to the probability information and the corresponding conclusion in the output result;

[0037] if no, generating a probability analysis report according to the output result.

[0038] In a second aspect, the present application provides an aircraft intention recognition probability estimation system, which comprises:

[0039] a basic information module, configured to acquire basic information of a target aircraft, and determine attribute information and state information in the basic information;

[0040] The antecedent structure module is configured to generate a semantic information fuzzy set as an antecedent structure parameter according to the attribute information and the state information.

[0041] The consequent structure module is configured to obtain target task information, and determine a membership function in a consequent structure according to the target task information.

[0042] The aggregation module is configured to combine the basic information and the target task information, and generate an intention probability prediction model of the target aircraft by aggregating preset semantic rules.

[0043] The intention judgment module is configured to obtain the basic information of the target aircraft at present, and determine a flight intention in combination with the intention probability prediction model.

[0044] In a third aspect, the present application provides a computer device, which comprises a memory and a processor, and the processor executes the method described above when running computer instructions stored in the memory.

[0045] In a fourth aspect, the present application provides a computer readable storage medium comprising instructions, which, when running on a computer, cause the computer to execute the method described above.

[0046] In summary, the present application has the following beneficial technical effects:

[0047] The present application obtains the basic information of the target aircraft, determines the attribute information and the state information in the basic information, generates a semantic information fuzzy set as an antecedent structure parameter according to the attribute information and the state information, obtains target task information, determines a membership function in a consequent structure according to the target task information, combines the basic information and the target task information, generates an intention probability prediction model of the target aircraft by aggregating preset semantic rules, obtains the basic information of the target aircraft at present, and determines a flight intention in combination with the intention probability prediction model. The complex nonlinear relationship is approximated by multiple rules, various complex data are fitted, each rule has semantic characteristics, and strong explainability is achieved, thereby achieving the technical effect of efficiently and accurately identifying the intention of the aircraft. BRIEF DESCRIPTION OF DRAWINGS

[0048] Figure 1 FIG. 1 is a computer device structure schematic diagram of a hardware running environment involved in an embodiment of the present application.

[0049] Figure 2 FIG. 2 is a flowchart of a first embodiment of an aircraft intention recognition probability estimation method of the present application.

[0050] Figure 3 FIG. 3 is a membership function diagram of a target conducting a reconnaissance action when the target is located in a general sea area in the first embodiment of the aircraft intention recognition probability estimation method of the present application.

[0051] Figure 4 is the membership function graph of the target conducting reconnaissance action when the target is located in the key sea area in the first embodiment of the aircraft intention identification probability estimation method of the present application;

[0052] Figure 5 is the membership function graph of the target conducting anti-submarine action when the target is located in the key sea area in the first embodiment of the aircraft intention identification probability estimation method of the present application;

[0053] Figure 6 is the membership function graph of the target conducting anti-submarine action when the target communicates with other targets in the first embodiment of the aircraft intention identification probability estimation method of the present application;

[0054] Figure 7 is the membership function graph of the target conducting anti-submarine action when the aircraft A approaches the target area in the first embodiment of the aircraft intention identification probability estimation method of the present application;

[0055] Figure 8 is the membership function graph of the target conducting anti-submarine action when the aircraft A approaches the target area in the first embodiment of the aircraft intention identification probability estimation method of the present application;

[0056] Figure 9 is the membership function graph of the target conducting anti-submarine action when the aircraft A approaches the target area in the first embodiment of the aircraft intention identification probability estimation method of the present application;

[0057] Figure 10 is the membership function graph of the target conducting reconnaissance action when the aircraft A approaches the target area in the first embodiment of the aircraft intention identification probability estimation method of the present application;

[0058] Figure 11 is the structural block diagram of the first embodiment of the aircraft intention identification probability estimation system of the present application. DETAILED DESCRIPTION

[0059] In order to make the purpose, technical scheme and advantages of the present application more clear, the present application is further described in detail below by means of the drawings and examples. It should be understood that the specific examples described herein are only used to explain the present application, and are not used to limit the present application.

[0060] Reference Figure 1 , Figure 1 is the computer device structural schematic diagram of the hardware running environment involved in the embodiment scheme of the present application.

[0061] As Figure 1As shown, the computer device may include: a processor 1001, such as a central processing unit (CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. The communication bus 1002 is used to enable communication between these components. The user interface 1003 may include a display screen and an input unit such as a keyboard; optionally, the user interface 1003 may also include a standard wired interface or a wireless interface. The network interface 1004 may optionally include a standard wired interface or a wireless interface (such as a Wireless-Fidelity (Wi-Fi) interface). The memory 1005 may be high-speed random access memory (RAM) or stable non-volatile memory (NVM), such as a disk storage device. The memory 1005 may also optionally be a storage device independent of the aforementioned processor 1001.

[0062] Those skilled in the art will understand that Figure 1 The structure shown does not constitute a limitation on the computer device and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0063] like Figure 1 As shown, the memory 1005, which serves as a storage medium, may include an operating system, a network communication module, a user interface module, and an aircraft intent recognition probability estimation program.

[0064] exist Figure 1 In the computer device shown, the network interface 1004 is mainly used for data communication with the network server; the user interface 1003 is mainly used for data interaction with the user; the processor 1001 and the memory 1005 in this application can be set in the computer device, and the computer device calls the aircraft intent recognition probability estimation program stored in the memory 1005 through the processor 1001 and executes the aircraft intent recognition probability estimation method provided in the embodiment of this application.

[0065] This application provides a method for probabilistic estimation of aircraft intent recognition, referring to... Figure 2 , Figure 2 This is a flowchart illustrating the first embodiment of the aircraft intent recognition probability estimation method of this application.

[0066] In this embodiment, the aircraft intent recognition probability estimation method includes the following steps:

[0067] Step S10: Obtain the basic information of the target aircraft, and determine attribute information and state information in the basic information.

[0068] It should be noted that the TS fuzzy rule model: Takagi-Sugeno (TS) fuzzy rule model, which is composed of a set of fuzzy rules describing the relationship between input variables and output variables. Each rule consists of an antecedent structure (IF) and a consequent structure (THEN), representing a local linear function. The output of the TS fuzzy rule model is the aggregation of the outputs of each rule, where the weights are determined by the degree of activation of the rule.

[0069] It can be understood that in the present embodiment, the TS fuzzy rule model is introduced. Because the algorithm design of intent recognition needs to explore an efficient and accurate explainable reasoning model under the premise of introducing domain knowledge. Since the fuzzy rule model can approximate complex nonlinear relationships with multiple rules, fit various complex data, and each rule has semantic features, it can have strong explainability.

[0070] In specific implementation, the present embodiment realizes explainable reasoning of aircraft target intent probability estimation based on fuzzy rules, explores the potential relationship between the attributes and behaviors of the aircraft equipment and the task, and constructs corresponding fuzzy rules to approximate complex nonlinear relationships. The aircraft intent recognition probability estimation method based on fuzzy rule reasoning can be divided into two key parts: one is the construction of fuzzy rules; the present embodiment considers all semantic information related to aircraft intent, i.e. the flight position, flight altitude, flight speed, distance from the target area, connectivity and distance from other equipment of the aircraft, and takes it as semantic knowledge in the antecedent structure of the rule. In the consequent structure, each local output is related to the antecedent structure, and the membership function of the corresponding fuzzy set is constructed. The second is to aggregate the constructed fuzzy rules to realize the final probability prediction, i.e. for different situations, output the probability of the corresponding aircraft target intent.

[0071] It can be understood that rules are an important way of knowledge representation in artificial intelligence, and rule-based models can capture the correlation between input and output spaces, reasonably divide complex problems into a series of sub-problems, and realize local processing of complex information. The model based on "IF-THEN" (if…, then…) rules can realize the modeling of complex nonlinear systems due to its robustness and interpretability. Through different algorithms and optimization methods, rule-based models can be constructed in various forms. Since fuzzy modeling can discover and mine hidden data information and realize human-centered knowledge representation, fuzzy rule systems have become a common framework for fuzzy rule generation and are widely used in many fields. Fuzzy rule modeling supports the basic paradigm of modularity, that is, each rule focuses on expressing the local modeling of a complex system, which can decompose a complex problem into a series of sub-problems. The outstanding advantage of fuzzy rule reasoning is that it can convert fuzzy concepts into scientific and feasible control strategies, effectively handling uncertainty and fuzziness in the real world. Mamdani fuzzy reasoning and Takagi-Sugeno (TS) fuzzy reasoning, as two representative types of fuzzy systems in fuzzy reasoning, have been widely used in various fields. Mamdani fuzzy model is a commonly used fuzzy reasoning method, which mainly includes three stages: fuzzification, fuzzy reasoning, and defuzzification. In the defuzzification stage, the fuzzy output is converted into an exact numerical result. The main difference between TS fuzzy reasoning and Mamdani fuzzy reasoning is that it combines the fuzzy reasoning and defuzzification stages into one stage. This means that the conclusion of TS fuzzy rules is directly numerical without the need for defuzzification.

[0072] TS fuzzy rules are generally composed of "If-Then" statements including antecedent and consequent structures. Fuzzy sets can provide interpretable representation and processing of knowledge in data, and conditions and consequent structures are usually represented in the form of fuzzy sets. That is, the general form of a rule can be represented as:

[0073] -If condition (antecedent structure) is Ai, Then conclusion (consequent structure) is Bi.

[0074] Where Ai and Bi are fuzzy sets corresponding to the spaces. Rules are a mathematical representation of the relationship between fuzzy sets, which can describe the local relationship between the input space and the output space. Rules have various forms and can express rich semantic information, providing powerful reasoning ability for fuzzy reasoning systems.

[0075] Therefore, the target intention prediction probability estimation model based on rule reasoning can be divided into two parts: one is the construction of fuzzy rules; in detail, in the antecedent structure, all semantic information related to the aircraft intention is considered in the embodiment, i.e. the flight position, flight height, flight speed, distance to the target area, connectivity and distance to other equipment, etc. of the equipment, and they are taken as semantic knowledge in the antecedent structure of the rule. In the consequent structure, each local output is related to the antecedent structure, and the membership function of the corresponding fuzzy set is constructed. Two is to aggregate the constructed fuzzy rules to realize the final probability prediction, i.e. for different cases, the probability of the corresponding aircraft intention is output.

[0076] It should be noted that the attribute information and state information obtained in the embodiment are as follows: the attribute information refers to the flight performance information corresponding to the target aircraft and the hardware size information of the aircraft itself; the state information refers to the flight position, flight height, flight speed, distance to the target area or distance to other equipment of the target aircraft. The embodiment does not limit the specific information content of the attribute information and state information.

[0077] Step S20: generating a semantic information fuzzy set as an antecedent structure parameter according to the attribute information and the state information.

[0078] It should be noted that the step of generating a semantic information fuzzy set as an antecedent structure parameter according to the attribute information and the state information includes: generating an attribute set according to the attribute information; generating a state set according to the state information; and generating a semantic information fuzzy set as an antecedent structure parameter according to the semantic information filtering strategy, the attribute set and the state set.

[0079] It should be noted that the step of generating a semantic information fuzzy set as an antecedent structure parameter according to the semantic information filtering strategy, the attribute set and the state set refers to obtaining semantic information in the attribute set and obtaining semantic information in the state set, and generating a semantic information fuzzy set by merging the semantic information in the attribute set and the semantic information in the state set. The semantic information fuzzy set is taken as an antecedent structure parameter to be applied in the TS fuzzy rule.

[0080] Step S30: obtaining target task information, and determining a membership function in a consequent structure according to the target task information.

[0081] It should be noted that the step of obtaining target task information and determining a membership function in a consequent structure according to the target task information includes: obtaining target task information and determining judgment conclusion information according to the target task information; determining whether the number of conclusions in the judgment conclusion information meets a preset conclusion condition; if yes, determining a membership function in a consequent structure according to the judgment conclusion information.

[0082] In specific implementation, the step of determining the membership function in the consequent structure based on the judgment conclusion information includes: determining the conclusion probability set in the judgment conclusion information; determining the target parameter type in the semantic information fuzzy set through semantic reasoning based on the conclusion probability set; and establishing the membership function in the consequent structure based on the target parameter type.

[0083] It should be noted that the step of establishing the membership function in the consequent structure based on the target parameter type includes: establishing a rule set of the semantic information fuzzy set based on the target parameter type; performing rationality screening in the rule set to generate a target rule set; and generating the membership function in the consequent structure by combining the target rule set with the rule aggregation method.

[0084] It is understood that the purpose of the preset conclusion condition in this embodiment is to ensure that there are two or more conclusions in the judgment conclusion information. If there is only one or fewer conclusions in the judgment conclusion information, the TS fuzzy rule used in this embodiment will not take effect, and the judgment result corresponding to the preset conclusion condition will be sent to the preset port for notification. Therefore, after setting the preset conclusion condition, the situation where the number of conclusions in the subsequent conclusions is less than two can be avoided, thus avoiding the waste of subsequent computing resources.

[0085] Step S40: Combine basic information and target mission information to generate a probability prediction model of the target aircraft's intent by aggregating preset semantic rules.

[0086] In a specific implementation, the aircraft intent recognition probability estimation method in this embodiment will be analyzed in detail, taking the aircraft intent recognition problem in a simple war game as an example.

[0087] Given scenario:

[0088] The areas near H Island, N Island Reef, the area directly opposite AA, and the area BB are all key sea areas.

[0089] The target area is within 12 nautical miles of H Island and near N Island Reef.

[0090] In practice, when equipment flies at an altitude of 5000-8000 meters, or 3000-4000 meters in key sea areas, it is very likely to be conducting reconnaissance operations; when the flight altitude is below 2500 meters, it is very likely to be conducting anti-submarine operations. For these scenarios, this embodiment uses fuzzy rules based on language descriptions to achieve semantic reasoning, enabling... Let represent the i-th rule related to the reconnaissance operation, and let its value represent the probability that it belongs to the reconnaissance operation; Let represent the j-th prediction rule related to anti-submarine operations, where its value represents the probability of conducting such operations. The process is as follows:

[0091] According to the flight position, flight height, flight speed, distance to the target area, and the communication and distance with other equipments, the membership functions of the reconnaissance and anti-submarine operations are determined.

[0092] There are two states of the aircraft position: general sea area and key sea area. In this part, the membership functions related to the flight height are constructed for the two states. For example, the flight height of aircraft A is between 5000-8000 meters, or between 3000-4000 meters in the key sea area, which is very likely to be reconnaissance operation, and the flight height below 2500 meters is very likely to be anti-submarine operation. Therefore, three rules are constructed for the two intention recognition results, that is:

[0093] Rule 1: IF aircraft A is in the general sea area, Then the probability of aircraft A performing reconnaissance operation (such as the membership function of the target performing reconnaissance operation as shown in FIG. 2) is represented as: Figure 3

[0094]

[0095] Rule 2: IF aircraft A is in the key sea area, Then the probability of aircraft A performing reconnaissance operation (such as the membership function of the target performing reconnaissance operation as shown in FIG. 3) is represented as: Figure 4

[0096]

[0097] Rule 3: IF aircraft A is in the key sea area, Then the probability of aircraft A performing anti-submarine operation (such as the membership function of the target performing anti-submarine operation as shown in FIG. 4) is represented as: Figure 5

[0098] In the above formula, h represents the flight height information of the aircraft.

[0099] In the case of communication between the aircraft and other aircraft, the membership functions related to the flight height are constructed. For example, when aircraft A has communication with aircraft B, and the flight height of aircraft A is between 1000-1500 meters, it is definitely reconnaissance operation. Therefore, the corresponding rule can be represented as:

[0100] Rule 4: IF aircraft A has communication with aircraft B, Then the probability of aircraft A performing anti-submarine operation (such as the membership function of the target performing anti-submarine operation as shown in FIG. 5) is represented as:

[0101] Figure 6

[0102] ​​​​

[0103] In practice, the distance between an aircraft and its target area can reveal its intentions.

[0104] Rule 5: If aircraft A approaches the target area, then the probability that the aircraft will conduct anti-submarine operations (e.g., ...) Figure 7 The membership function of the target's anti-submarine operations when aircraft A approaches the target area can be expressed as:

[0105]

[0106] Rule 6: If aircraft A approaches the target area, then the probability that the aircraft will conduct anti-submarine operations (e.g., ...) Figure 8 The membership function of the target's anti-submarine operations when aircraft A approaches the target area can be expressed as:

[0107]

[0108] In the above formula, d represents the aircraft's flight altitude information.

[0109] When analyzing an aircraft's intentions, its flight speed and distance from other aircraft are considered. For example, if aircraft A is 150-300 kilometers outside of aircraft B, and carrier-based aircraft are taking off from aircraft B, aircraft A is likely conducting reconnaissance operations. If aircraft A is outside of aircraft B and less than 150 kilometers away, it is more likely that aircraft A is conducting anti-submarine warfare operations. The corresponding rules can be expressed as follows:

[0110] Rule 7: If aircraft A and aircraft B are close to each other at a distance of d km, and a carrier-based aircraft takes off from aircraft B, then the probability that aircraft A will conduct anti-submarine operations is expressed as (e.g., Figure 9 As shown):

[0111]

[0112] Rule 8: If aircraft A and aircraft B are close together and their speed is v km / h, and aircraft B has a carrier-based aircraft taking off, then the probability that aircraft A will conduct a reconnaissance operation is (e.g., ...). Figure 10 The membership function of a target conducting reconnaissance operations when aircraft A and aircraft B are close together, and carrier-based aircraft take off from aircraft B, is expressed as follows:

[0113]

[0114] The field knowledge based on the experience of experts and the general rules of the problem are comprehensively considered, and 8 rules are constructed around the reconnaissance task and the anti-submarine task. Four of them are related to the reconnaissance action, and the other four are related to the reconnaissance action. On this basis, the intention probability of the aircraft, i.e. X probability and Y probability, is obtained through rule aggregation.

[0115] The final predicted X probability can be represented as p 吊撞 :

[0116]

[0117] The final predicted Y probability can be represented as p 链 :

[0118]

[0119] Step S50: Obtain the basic information of the current target aircraft and determine the flight intention in combination with the intention probability prediction model.

[0120] Further, in order to improve the accuracy of flight intention determination, the step of obtaining the basic information of the current target aircraft and determining the flight intention in combination with the intention probability prediction model includes: obtaining the basic information of the current target aircraft to generate an input parameter set; inputting the input parameter set into the intention probability prediction model and obtaining an output result; obtaining a conclusion and corresponding probability information in the output result; determining the flight intention according to the conclusion and the corresponding probability information.

[0121] It should be noted that the step of determining the flight intention according to the conclusion and the corresponding probability information includes: obtaining a preset determination threshold, and determining whether there is probability information that meets the preset determination threshold in the output result; if yes, determining the flight intention according to the probability information and the corresponding conclusion in the output result; if no, generating a probability analysis report according to the output result.

[0122] It can be understood that the preset determination threshold can be filtered according to specific probability information in this embodiment. The setting of the preset determination threshold is pre-set according to specific use conditions.

[0123] In specific implementation, in order to verify the effectiveness of the method for target intention recognition, data in three types of effective scenes are collected on a war game platform for verification.

[0124] Given the background: H island, N island reef, AA area, and BB area are all key sea areas; 12 nautical miles around H island, and target area around N island reef;

[0125] Scenario one: aircraft A now flies at 3500 meters height in the vicinity of H island 30 nautical miles, and the same aircraft B distance of about 150 kilometers, the possible action of aircraft A is?

[0126] According to the condition "H island near 30 nautical miles", the position of equipment A is judged as the key sea area. According to the flight height of A and the distance from B, the 8 rules are used in turn to judge, which can be obtained:

[0127]

[0128]

[0129] Therefore, the probability of aircraft A's intention is 0.75 for reconnaissance action and 0.15 for anti submarine action, and the aircraft may perform the task of reconnaissance action.

[0130] Scenario two: aircraft A at 420 km / h in the vicinity of H island, and in the periphery of aircraft B 200 kilometers, and B has aircraft taking off, the possible action of aircraft A is?

[0131] According to the condition "H island near 30 nautical miles", the position of equipment A is judged as the key sea area. At the same time, the aircraft A is close to the aircraft B, and the aircraft B has aircraft taking off, according to the distance d = 200 kilometers, the rule 7 can be calculated At the same time, the flight speed v = 420 km / h, the rule 8 can be calculated The final rule consequent structure and aggregation result are as follows:

[0132]

[0133] Therefore, the probability of aircraft A's intention is 0.28 for reconnaissance action and 0.6 for anti submarine action, and the aircraft A may perform the task of anti submarine action.

[0134] Scenario three: aircraft A flies in the vicinity of N island reef, and the flight height is 4000 meters, the distance from the target area is 200 kilometers, at this time the possible action of aircraft A is?

[0135] According to the condition "aircraft A flies in the vicinity of N island reef", the position of aircraft A is judged as the key sea area. At the same time, the flight height h = 2500 meters, according to the rule 2 and the rule 3, it can be calculated that The aircraft approaches the target area and the distance is 200 kilometers, according to the rule 5 and the rule 6, it can be calculated that The final rule consequent structure and aggregation result are as follows

[0136]

[0137]

[0138] Therefore, the probability of the aircraft intending to be the reconnaissance action is 0.665, the probability of the anti-submarine action is 0.5, and the task that the aircraft can perform is the reconnaissance action.

[0139] It should be noted that, compared with the prior art, the embodiment has the following distinguishing points:

[0140] 1) Analyzing the properties and behaviors of the target equipment, considering all semantic information related to the target aircraft intention, generating a fuzzy set pair to represent and process knowledge in an interpretable manner;

[0141] 2) Effectively utilizing experience information to construct semantic fuzzy rules to describe the local relationship of aircraft target intention probability estimation;

[0142] 3) Based on the constructed semantic rules, the intention probability prediction is realized by an aggregation method, which is beneficial to the balance between the accuracy and interpretability of the intention prediction model.

[0143] The embodiment obtains the basic information of the target aircraft, determines the attribute information and state information in the basic information, generates a semantic information fuzzy set as the antecedent structure parameter according to the attribute information and the state information, obtains target task information, determines a second fuzzy set as the subsequent conclusion according to the target task information, determines the membership function between the semantic information fuzzy set and the second fuzzy set according to the semantic fuzzy rule strategy combined with the basic information and the target task information, obtains the basic information of the current target aircraft, and determines the flight intention of the target aircraft according to the membership function. By approximating complex nonlinear relationships through multiple rules, fitting various complex data, and each rule having semantic characteristics, strong interpretability can be achieved, and the technical effect of efficiently and accurately identifying aircraft intention is realized.

[0144] In addition, the embodiment of the application also provides a computer readable storage medium, and the storage medium stores an aircraft intention identification probability estimation program. When the aircraft intention identification probability estimation program is executed by a processor, the steps of the aircraft intention identification probability estimation method are realized.

[0145] Reference Figure 11 , Figure 11 is a structural block diagram of the first embodiment of the aircraft intention identification probability estimation system of the application.

[0146] As Figure 11 shown, the aircraft intention identification probability estimation system provided by the embodiment of the application comprises:

[0147] The basic information module 10 is configured to obtain the basic information of the target aircraft, and determine the attribute information and the state information in the basic information.

[0148] The antecedent structure module 20 is configured to generate a semantic information fuzzy set as an antecedent structure parameter according to the attribute information and the state information.

[0149] The consequent structure module 30 is configured to obtain target task information, and determine a membership function in the consequent structure according to the target task information.

[0150] The aggregation module 40 is configured to combine the basic information and the target task information, and generate an intention probability prediction model of the target aircraft by aggregating preset semantic rules.

[0151] The intention judgment module 50 is configured to obtain the basic information of the current target aircraft, and determine a flight intention in combination with the intention probability prediction model.

[0152] It should be understood that the above is only an example, and does not constitute any limitation on the technical solutions of the present application. In specific applications, those skilled in the art can make settings as needed, and the present application does not limit this.

[0153] The embodiment obtains the basic information of the target aircraft, determines the attribute information and the state information in the basic information, generates a semantic information fuzzy set as an antecedent structure parameter according to the attribute information and the state information, obtains target task information, and determines a membership function in the consequent structure according to the target task information. The basic information and the target task information are combined, and an intention probability prediction model of the target aircraft is generated by aggregating preset semantic rules. The basic information of the current target aircraft is obtained, and a flight intention is determined in combination with the intention probability prediction model. The complex nonlinear relationship is approximated by a plurality of rules, a plurality of complex data are fitted, each rule has semantic characteristics, and strong interpretability can be achieved. The technical effect of efficiently and accurately identifying the intention of the aircraft is achieved.

[0154] In an embodiment, the antecedent structure module 20 is further configured to generate an attribute set according to the attribute information, generate a state set according to the state information, and generate a semantic information fuzzy set as an antecedent structure parameter in combination with the attribute set and the state set according to a semantic information screening strategy.

[0155] In an embodiment, the consequent structure module 30 is further configured to obtain target task information and determine a judgment conclusion information according to the target task information, determine whether the number of conclusions in the judgment conclusion information meets a preset conclusion condition, and if so, determine a membership function in the consequent structure according to the judgment conclusion information.

[0156] In an embodiment, the consequent structure module 30 is further configured to determine a conclusion probability set in the judgment conclusion information, determine a target parameter type in the semantic information fuzzy set by a semantic reasoning manner according to the conclusion probability set, and establish a membership function in the consequent structure according to the target parameter type.

[0157] In an embodiment, the back structure module 30 is further configured to establish a rule set of the semantic information fuzzy set according to the target parameter type; perform rationality screening in the rule set to generate a target rule set; and generate the membership function in the back structure by combining the target rule set in a rule aggregation manner.

[0158] In an embodiment, the intention judgment module 50 is further configured to obtain basic information of the current target aircraft to generate an input parameter set; input the input parameter set into the intention probability prediction model and obtain an output result; obtain a conclusion and corresponding probability information in the output result; and determine the flight intention according to the conclusion and the corresponding probability information.

[0159] In an embodiment, the intention judgment module 50 is further configured to obtain a preset determination threshold, and determine whether there is probability information satisfying the preset determination threshold in the output result; if yes, determine the flight intention according to the probability information and the corresponding conclusion in the output result; and if no, generate a probability analysis report according to the output result.

[0160] It should be noted that the above-described workflow is merely illustrative and does not limit the protection scope of the present application. In actual application, a person skilled in the art can select part or all of the above-described workflow to achieve the purpose of the embodiment, and the selection is not limited herein.

[0161] In addition, technical details not described in detail in the embodiment can be referred to the method for estimating the probability of aircraft intention recognition provided by any embodiment of the present application, and will not be described herein.

[0162] In addition, it should be noted that in this document, the terms "include", "contain" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or system including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or system. Without more limitations, the element defined by the statement "including a" does not exclude the presence of another identical element in the process, method, article or system including the element.

[0163] The above-mentioned serial numbers of the embodiments of the present application are only for description, and do not represent the advantages and disadvantages of the embodiments.

[0164] Those skilled in the art can clearly understand the above-mentioned embodiment method can be realized by means of software and the necessary general hardware platform, of course, can also be through hardware, but in many cases the former is a better embodiment. Based on such understanding, the technical solutions of the present application essentially or say the part of the prior art contribution can be embodied in the form of software products, the computer software product is stored in a storage medium (such as read only memory (Read Only Memory, ROM) / RAM, magnetic disk, optical disc), including a number of instructions to make a terminal device (may be a mobile phone, computer, server, or network equipment, etc.) executes the method of each embodiment of the present application.

[0165] The above is only the preferred embodiment of the present application, not therefore limit the patent scope of the present application, all the equivalent structure or equivalent process transformation using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. An aircraft intent recognition probability estimation method, characterized in that, The method comprises the following steps: acquiring basic information of a target aircraft, determining attribute information and state information in the basic information; generating a semantic information fuzzy set as a premise structure parameter according to the attribute information and the state information; acquiring target task information, and determining a membership function in a consequent structure according to the target task information; combining the basic information and the target task information, and generating an intention probability prediction model of the target aircraft by aggregating preset semantic rules; acquiring current basic information of the target aircraft and determining a flight intention by combining the intention probability prediction model; The step of acquiring target task information and determining a membership function in a consequent structure according to the target task information comprises the following steps: acquiring target task information and determining judgment conclusion information according to the target task information; determining whether the number of conclusions in the judgment conclusion information meets a preset conclusion condition; if yes, determining a membership function in a consequent structure according to the judgment conclusion information; The step of determining a membership function in a consequent structure according to the judgment conclusion information comprises the following steps: determining a conclusion probability set in the judgment conclusion information; determining a target parameter type in the semantic information fuzzy set by a semantic reasoning manner according to the conclusion probability set; establishing a membership function in a consequent structure according to the target parameter type; The step of establishing a membership function in a consequent structure according to the target parameter type comprises the following steps: establishing a rule set of the semantic information fuzzy set according to the target parameter type; performing rationality screening in the rule set to generate a target rule set; generating a membership function in a consequent structure by combining the target rule set in a rule aggregation manner.

2. The aircraft intent recognition probability estimation method of claim 1, wherein, The step of generating a semantic information fuzzy set as a premise structure parameter according to the attribute information and the state information comprises the following steps: generating an attribute set according to the attribute information; generating a state set according to the state information; generating a semantic information fuzzy set as a premise structure parameter by combining the attribute set and the state set according to a semantic information screening strategy.

3. The aircraft intent recognition probability estimation method of claim 1, wherein, The step of acquiring current basic information of the target aircraft and determining a flight intention by combining the intention probability prediction model comprises the following steps: acquiring current basic information of the target aircraft to generate an input parameter set; inputting the input parameter set into the intention probability prediction model and acquiring an output result; acquiring a conclusion and corresponding probability information in the output result; determining a flight intention according to the conclusion and the corresponding probability information.

4. The aircraft intent recognition probability estimation method of claim 3, wherein, The step of determining a flight intention according to the conclusion and the corresponding probability information comprises the following steps: acquiring a preset determination threshold, and determining whether there is probability information meeting the preset determination threshold in the output result; if yes, determining a flight intention according to the probability information and the corresponding conclusion in the output result; if no, generating a probability analysis report according to the output result.

5. An aircraft intent recognition probability estimation system characterized by, The aircraft intention recognition probability estimation system comprises the following steps: A basic information module is configured to acquire basic information of a target aircraft, and determine attribute information and state information in the basic information; A premise structure module is configured to generate a semantic information fuzzy set as a premise structure parameter according to the attribute information and the state information; A consequent structure module is configured to acquire target task information, and determine a membership function in a consequent structure according to the target task information; An aggregation module is configured to combine the basic information and the target task information, and generate an intention probability prediction model of the target aircraft by aggregating a preset semantic rule; An intention judgment module is configured to acquire current basic information of the target aircraft, and determine a flight intention in combination with the intention probability prediction model.

6. A computer device, comprising: The device comprises a memory and a processor, and the processor executes the method according to any one of claims 1 to 4 when running computer instructions stored in the memory.

7. A computer-readable storage medium, characterized in that, The instructions, when running on a computer, cause the computer to execute the method according to any one of claims 1 to 4.

Images