Unmanned Aerial Vehicle Combat Performance Evaluation Method, Evaluation System, Memory, and Electronic Device

Abstract

The present invention provides a method and a system for evaluating the combat performance of an unmanned aerial vehicle (UAV), a memory, and an electronic device, belonging to the technical field of UAVs. The method includes: constructing a Bayesian network; improving the Bayesian network to form a weighted Bayesian algorithm; using the Euclidean distance metric algorithm and the weighted Bayesian algorithm to determine the strike effectiveness value of the UAV against a target at the current moment; constructing a weighted dynamic Bayesian algorithm based on the weighted Bayesian algorithm, and determining the strike effectiveness value between the UAV and its matched target according to the strike effectiveness value of the UAV against the target at the current moment and the weighted dynamic Bayesian algorithm; determining the strike effectiveness values between all UAVs and their matched targets to form a strike effectiveness data set; and determining the combat performance evaluation value of the UAV based on the strike effectiveness data set and the matching rate, so as to evaluate the combat performance. The present invention improves the correlation between multiple evaluation indexes of the UAV system and can more accurately reflect the combat ability of the UAV system.

Description

Technical Field

[0001] The present invention belongs to the technical field of unmanned aerial vehicles (UAVs), and relates to the combat performance evaluation technology of UAVs. Specifically, it is a method, a system, a memory, and an electronic device for evaluating the combat performance of UAVs. Background Art

[0002] UAVs have become crucial in performing precision strikes and attack decisions. An efficient and accurate strike evaluation and decision support system plays a pivotal role in improving the overall combat ability of UAVs, reducing potential risks, rationally allocating resources, and coping with various complex environments.

[0003] When UAVs conduct strike drills during actual combat drills, it is necessary to evaluate the effect of the comprehensive algorithm on the strike target. In essence, it is to evaluate the technical performance of UAVs when performing specific tasks. The evaluation process is not only a test of a single technology, but also a comprehensive performance test of multiple algorithms such as target recognition, continuous tracking, and rapid decision-making. In the environment of actual combat drills, the evaluation using the comprehensive algorithm can more comprehensively reflect the actual combat ability of the UAV system. The core purpose of evaluating the comprehensive algorithm is to detect the performance at the technical level of UAVs, aiming to continuously optimize the combat performance of UAVs. Through data collection and analysis in actual combat drills, it is possible to more accurately understand the performance of UAVs in various complex environments and the effects of various algorithms in the UAV system in actual applications. This not only helps to discover problems existing in the UAV system, but also helps to provide valuable data support for the subsequent improvement and enhancement of UAVs.

[0004] Currently, when evaluating the comprehensive algorithm of the UAV system, multiple evaluation indicators are independent of each other. For example, for the perception ability of UAVs, it is only evaluated based on the positioning accuracy of various sensors such as GPS, inertial measurement unit (IMU), camera, and lidar carried by the UAV, without considering the correlation between other indicators and the perception ability. However, in actual combat drills, there are correlations and mutual influences among multiple evaluation indicators of UAVs. For example, the perception ability of UAVs will affect their tracking ability. Therefore, the data provided by the existing evaluation algorithms is inaccurate and cannot provide strong data support for the improvement of the overall performance of UAVs. Summary of the Invention

[0005] In view of the technical problem described in the above background art that multiple evaluation indicators are independent of each other during the evaluation of the existing evaluation algorithms and cannot provide strong data support for the improvement of the overall performance of UAVs, the present invention proposes a method, a system, a memory, and an electronic device for evaluating the combat performance of UAVs.

[0006] The present invention constructs a Bayesian network based on multiple evaluation indicators, and at the same time uses the grey relational algorithm to improve the Bayesian network, weakening the independence assumption of multiple evaluation indicators, making multiple evaluation indicators correlated with each other, fully considering the correlation between multiple evaluation indicators and the volatility of multiple evaluation indicator data. Then, the strike benefit value of the UAV against the target is determined, and finally the strike benefit value between each UAV and its corresponding target is determined to form a strike benefit data set. Combining the strike benefit data set and the matching rate, the combat performance evaluation value of the UAV is determined, so as to evaluate the combat performance of the UAV. It improves the correlation between multiple evaluation indicators of the UAV system, can provide strong data support for the improvement of the overall performance of the UAV, and thus more accurately reflects the combat ability of the UAV system.

[0007] To solve the above technical problems, the present invention adopts the following technical solutions:

[0008] The UAV combat performance evaluation method of the present invention includes the following steps:

[0009] Determine multiple evaluation indicators of the UAV combat performance, and construct a Bayesian network according to the correlation relationship between the multiple evaluation indicators;

[0010] Use the grey relational algorithm to improve the Bayesian network to form a weighted Bayesian algorithm;

[0011] Use the Euclidean distance metric algorithm and the weighted Bayesian algorithm to determine the strike benefit value of the UAV against the target at the current moment; construct a weighted dynamic Bayesian algorithm based on the weighted Bayesian algorithm, and determine the strike benefit value between the UAV and its matching target according to the strike benefit value of the UAV against the target at the current moment and the weighted dynamic Bayesian algorithm;

[0012] Repeat the above steps until the strike benefit values between all UAVs and their matching targets are determined to form a strike benefit data set;

[0013] Determine the matching rate between the UAV and the target;

[0014] Based on the strike benefit data set and the matching rate, determine the combat performance evaluation value of the UAV, and evaluate the combat performance of the UAV according to the combat performance evaluation value of the UAV.

[0015] Further defined, the use of the grey relational algorithm to improve the Bayesian network to form a weighted Bayesian algorithm specifically includes:

[0016] Use the grey relational algorithm to assign different weights to each node of the Bayesian network, and combine the different weights of each node to determine the weight of each sub-node relative to the root node in the Bayesian network, so as to form a weighted Bayesian algorithm.

[0017] Further limitation: The specific process of determining the strike effectiveness value of the UAV against the target at the current moment using the Euclidean distance metric algorithm and the weighted Bayesian algorithm includes: using the Euclidean distance metric algorithm to compare and judge the similarity of evaluation indicators at different moments, and determining the strike effectiveness value of the UAV against the target at the current moment according to the similarity judgment result of evaluation indicators at different moments or the weighted Bayesian algorithm.

[0018] Further limitation: The specific process of using the Euclidean distance metric algorithm to compare and judge the similarity of evaluation indicators at different moments, and determining the strike effectiveness value of the UAV against the target at the current moment according to the similarity judgment result of evaluation indicators at different moments or the weighted Bayesian algorithm includes:

[0019] Using the Euclidean distance metric algorithm to judge the similarity of evaluation indicators corresponding to two adjacent moments. If the similarity is greater than the similarity threshold, use the strike effectiveness value of the previous moment as the strike effectiveness value of the UAV against the target at the current moment; if the similarity does not exceed the similarity threshold, determine the strike effectiveness value of the UAV against the target at the current moment according to the weighted Bayesian algorithm.

[0020] Further limitation: The specific process of constructing the weighted dynamic Bayesian algorithm based on the weighted Bayesian algorithm includes: adding time series data to the weighted Bayesian algorithm to construct the weighted dynamic Bayesian algorithm.

[0021] Further limitation: The specific process of determining the matching rate between the UAV and the target includes:

[0022] Determining the threat level of the target, matching the UAV and the target according to the threat level of the target and the strike effectiveness value between the UAV and its matching target, and determining the matching rate.

[0023] Further limitation: The evaluation indicators include strike effectiveness, sensing ability, tracking ability, maneuverability, and real-time performance; among them, the sensing ability includes false alarm rate, missed detection rate, and positioning accuracy, and the maneuverability includes relative speed, relative altitude, and heading angle.

[0024] The UAV combat performance evaluation system of the present invention is applied to the above-mentioned UAV combat performance evaluation method, and includes:

[0025] A Bayesian network construction module, used to determine multiple evaluation indicators of the UAV combat performance, and construct a Bayesian network according to the correlation relationship between the multiple evaluation indicators;

[0026] An improvement module: used to improve the Bayesian network using the grey relational algorithm to form a weighted Bayesian algorithm;

[0027] Strike benefit value determination module: used to determine the strike benefit value of the UAV against the target at the current moment using the Euclidean distance metric algorithm and the weighted Bayesian algorithm; used to construct a weighted dynamic Bayesian algorithm based on the weighted Bayesian algorithm, and determine the strike benefit value between the UAV and its matching target according to the strike benefit value of the UAV against the target at the current moment and the weighted dynamic Bayesian algorithm;

[0028] Strike benefit data set determination module: used to repeat the improvement module and the strike benefit value determination module until the strike benefit values between all UAVs and their matching targets are determined, forming a strike benefit data set;

[0029] Matching module: used to determine the matching rate between the UAV and the target;

[0030] And the combat performance evaluation module: used to determine the combat performance evaluation value of the UAV based on the strike benefit data set and the matching rate, and evaluate the combat performance of the UAV according to the combat performance evaluation value of the UAV.

[0031] The memory of the present invention stores program files, and the program files are executed to implement the program instructions formed by the above UAV combat performance evaluation method.

[0032] The electronic device of the present invention includes a processor and a memory that are coupled to each other, wherein,

[0033] The memory: used to store the program instructions formed by the above UAV combat performance evaluation method;

[0034] The processor: used to execute the program instructions stored in the memory.

[0035] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0036] 1. The UAV combat performance evaluation method of the present invention constructs a Bayesian network based on multiple evaluation indicators, and at the same time uses the grey relational algorithm to improve the Bayesian network, weakening the independence assumption of multiple evaluation indicators, making multiple evaluation indicators related to each other, fully considering the correlation between multiple evaluation indicators and the volatility of multiple evaluation indicator data, then determining the strike benefit value of the UAV against the target, and finally determining the strike benefit value between each UAV and its corresponding target, forming a strike benefit data set, and then combining the strike benefit data set and the matching rate to determine the combat performance evaluation value of the UAV, so as to evaluate the combat performance of the UAV. It improves the correlation between multiple evaluation indicators executed by the UAV, can provide strong data support for the improvement of the overall performance of the UAV, and thus more accurately reflects the combat ability of the UAV system.

[0037] 2. In the present invention, the grey relational algorithm is first used to assign different weights to each node of the Bayesian network, so as to differentially assign different weights to the evaluation indexes of each node in the Bayesian network. Then, the weights of each child node in the Bayesian network relative to the root node are determined by combining the different weights of each node, realizing the objective assignment of the weights of multiple evaluation indexes, effectively enhancing the mutual dependence relationship between each node, weakening the independence assumption in the existing comprehensive algorithm, and thus improving the evaluation performance.

[0038] 3. In the present invention, the Euclidean distance metric algorithm is used to compare and judge the similarity of multiple evaluation indexes corresponding to each moment, so as to reduce the computational complexity and improve the real-time performance of the dynamic Bayesian algorithm. In addition, in an environment with limited computing resources, it can also effectively save the consumption of computing resources.

[0039] 4. Using the UAV combat performance evaluation method of the present invention to evaluate the comprehensive performance of UAVs in actual combat drills can accumulate a large amount of combat performance data of UAVs. Through the combat performance data, not only can it provide a basis for the current performance optimization of UAVs, but also it can provide strong support for the research and improvement of future UAV technologies. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 is a schematic diagram of the UAV combat performance evaluation method of the present invention;

[0041] Figure 2 is a schematic diagram of the UAV combat performance evaluation system of the present invention;

[0042] Figure 3 is a schematic diagram of the Bayesian network;

[0043] Figure 4 is a comparison chart of the evaluation results formed by the UAV combat performance evaluation method of the present invention and the existing comprehensive algorithm. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0044] The technical solution of the present invention will be further explained below with reference to the drawings and embodiments, but the present invention is not limited to the following embodiments.

[0045] See Figure 1 , the present invention provides a UAV combat performance evaluation method, including the following steps:

[0046] S1: Determine multiple evaluation indexes of the UAV combat performance, and construct a Bayesian network according to the correlation relationship between the multiple evaluation indexes;

[0047] See Figure 3, in the present invention, the evaluation indicators include strike effectiveness, perception ability, tracking ability, maneuverability, and real-time performance. The perception ability is a key indicator to measure whether a drone can effectively identify and locate a target, including false alarm rate, miss detection rate, and positioning accuracy. The false alarm rate is the proportion of misjudging non-targets as targets. The miss detection rate is the proportion of failing to detect real targets. The positioning accuracy is the accuracy of obtaining the target position through the relationship between the target image and coordinates. The tracking ability is determined by the number of times the target is lost during the tracking process. If the drone frequently loses the target, the probability of successful strike will be significantly reduced. The maneuverability is an important indicator to measure the flexibility and sensitivity of the drone during movement, including relative speed, relative height, and heading angle. The real-time performance is a measure of the reaction speed of the drone. A highly sensitive reaction speed can ensure the reaction time of the drone from receiving the control signal to the actual action, which affects the strike effect of the drone.

[0048] Construct a Bayesian network based on the correlation relationships (causal relationships and logical dependencies) between the evaluation indicators, and at the same time determine the node variables of the Bayesian network. The Bayesian network has a total of 11 node variables. Among them, the false alarm rate, miss detection rate, positioning accuracy, relative speed, relative height, and heading angle are observable node variables, and the strike effectiveness, perception ability, tracking ability, maneuverability, and real-time performance are hidden node variables, as shown in Table 1.

[0049] Table 1: Node Variable Table

[0050]

[0051] S2: Use the grey relational algorithm to improve the Bayesian network to form a weighted Bayesian algorithm;

[0052] Step S2 is specifically as follows: Use the grey relational algorithm to assign different weights to each node of the Bayesian network, and determine the weights of each sub-node relative to the root node in the Bayesian network according to the different weights of each node, so as to form a weighted Bayesian algorithm to determine the relative importance of the evaluation indicators to the strike effectiveness.

[0053] Specifically, calculate the weights of the sub-nodes (false alarm rate, miss detection rate, and positioning accuracy) relative to the root node (perception ability) respectively; calculate the weights of the sub-nodes (relative speed, relative height, and heading angle) relative to the root node (maneuverability) respectively; calculate the weights of the sub-nodes (perception ability, tracking ability, maneuverability, and real-time performance) relative to the root node (strike effectiveness) respectively.

[0054] Reference sequence: Select the maximum and minimum values of the sub-nodes as the reference sequences. Suppose there are n evaluation indicators, then two reference sequences can be obtained, namely the maximum value reference sequence and the minimum value reference sequence, where the maximum value reference sequence Minimum value reference sequence i is the serial number of the evaluation index, and n is the number of evaluation indexes. is the maximum reference value of the i-th evaluation index. is the minimum reference value of the i-th evaluation index.

[0055] Evaluation sequence: It is composed of the actual values of the evaluation indexes at different times. The evaluation sequence where j is the time value.

[0056] In order to eliminate the influence of the dimensions and orders of magnitude of different evaluation indexes and make the evaluation indexes comparable, the range method is used to standardize the evaluation sequence. The standardized sequence is is the dimensionless value of the i-th evaluation index at the j-th time, and its calculation formula is:

[0057]

[0058] In the formula, is the actual value of the i-th evaluation index at the j-th time, and min(X j ) is the minimum value at the j-th time in the evaluation sequence; max(X j ) is the maximum value at the j-th time in the evaluation sequence.

[0059] The grey relational algorithm is used to balance the correlation degree between the evaluation sequence and the reference sequence, which is represented by the maximum correlation coefficient of the i-th evaluation index at the j-th time and the minimum correlation coefficient of the i-th evaluation index at the j-th time respectively:

[0060]

[0061] In the formula, ρ is the resolution coefficient, and its value is 0.5; is the minimum difference between two levels; is the maximum difference between two levels.

[0062] The grey relational depth coefficient λ ij is:

[0063]

[0064] For the average correlation degree μ i of each evaluation index on all evaluation objects and the standard deviation σ i of the correlation degree are:

[0065]

[0066] In the formula, m is the number of evaluation sequences; other parameter definitions are the same as above.

[0067] According to the coefficient of variation method, the weight of each evaluation index is equal to its coefficient of variation divided by the sum of the coefficients of variation of all evaluation indexes, and then normalized to finally obtain the weight ω of the evaluation index i , and the calculation formula is as follows:

[0068]

[0069] In the formula, S i is the coefficient of variation of the i-th evaluation index; ω i is the weight of the i-th evaluation index; the definitions of other parameters are the same as above.

[0070] The present invention uses the grey relational algorithm and the coefficient of variation method to determine the weights of the evaluation indexes, comprehensively considering the correlation between multiple evaluation indexes and the volatility of data, making the weight allocation of multiple evaluation indexes more objective.

[0071] S3: Use the Euclidean distance metric algorithm and the weighted Bayesian algorithm to determine the strike effectiveness value of the UAV on the target at the current moment; construct a weighted dynamic Bayesian algorithm based on the weighted Bayesian algorithm, and determine the strike effectiveness value between the UAV and its matching target according to the strike effectiveness value of the UAV on the target at the current moment and the weighted dynamic Bayesian algorithm.

[0072] Step S3 is specifically: use the Euclidean distance metric algorithm to compare and judge the similarity of evaluation indexes at different moments, and determine the strike effectiveness value of the UAV on the target at the current moment according to the similarity judgment result of evaluation indexes at different moments or the weighted Bayesian algorithm. Specifically, use the Euclidean distance metric algorithm to judge the similarity of evaluation indexes corresponding to two adjacent moments. If the similarity is greater than the similarity threshold, use the strike effectiveness value of the previous moment as the strike effectiveness value of the UAV on the target at the current moment; if the similarity does not exceed the similarity threshold, determine the strike effectiveness value of the UAV on the target at the current moment according to the weighted Bayesian algorithm.

[0073] Set the evaluation sequence of the evaluation index at the (j - 1) moment as The evaluation sequence of the evaluation index at the j moment is The weight vector ω of the evaluation index = ω1, ω2, ω3,..., ω i ,..., ω n , n is the number of evaluation indexes, and calculate the similarity between X j-1 and X j through the Euclidean distance metric algorithm.

[0074] Set the similarity threshold δ. If the similarity of the evaluation indicators corresponding to two adjacent moments is greater than the similarity threshold, the strike benefit value at the previous moment is used as the strike benefit value of the UAV against the target at the current moment; if the similarity of the evaluation indicators corresponding to two adjacent moments does not exceed the similarity threshold, the strike benefit value of the UAV against the target at the current moment is determined according to the weighted Bayesian algorithm.

[0075] The weighted Bayesian algorithm weakens the independence assumption of evaluation indicators through weighting, and can more accurately evaluate the UAV strike benefit at each moment.

[0076] (1) A Bayesian network is a directed acyclic graph that represents the probabilistic dependence relationship between variables. The conditional independence assumption implied in the Bayesian network holds that the child nodes in the background network of the conditional nodes are all independent of each other. However, in practical applications, due to the widespread correlation among multiple evaluation indicators, the independence assumption needs to be adjusted. To improve the overall evaluation efficiency, the connection between nodes is enhanced by assigning different weights to each node variable, thereby weakening the existing independence assumption and achieving a more accurate and comprehensive strike benefit evaluation. Among them, the weighted Bayesian algorithm is as follows:

[0077]

[0078] In the formula, v k is the value state of the kth root node; k is the number of root nodes; V l is the value state of the lth child node, and l is the number of child nodes; the posterior probability P(v k ) is derived from the prior probability distribution P(v i ), the conditional probability distribution P(V k |v i ) of the child node, and the weight ω k of the ith evaluation; (V1, V2,..., V l ) is the set of child nodes of the root node v l . k

[0079] (2) Based on the weighted Bayesian algorithm, the weighted Bayesian algorithm is combined with time series data to form a dynamic Bayesian network. The state transition probability between two moments is set in the weighted dynamic Bayesian algorithm, that is, the prior probability of the strike benefit is dynamically updated at each moment. Let P(C[j + 1]C[j]) be the state transition probability matrix; P(C[j] ′ ) be the posterior probability of the strike benefit node at the current moment, and the prior probability P(C[j + 1]) of the strike benefit at the next moment can be expressed as:

[0080] P(C[j + 1]) = P(C[j]′ )P(C[j + 1]C[j])

[0081] Wherein, the posterior probability P(C[j] ′ ) of the strike benefit node at the current moment is equal to the posterior probability P(v k |V1, V2,..., V l ) of the current strike benefit.

[0082] S4: Determine the matching rate between the UAV and the target;

[0083] Step S4 is specifically as follows: Determine the threat level of the target, and match the UAV with the target according to the threat level of the target and the strike benefit value between the UAV and its matched target, and determine the matching rate. Specifically, one UAV corresponds to one target and one strike benefit value; that is, the correspondence between multiple UAVs and multiple targets is: each UAV corresponds to the strike benefit value of one target.

[0084] S5: Determine the combat performance evaluation value of the UAV based on the strike benefit data set and the matching rate, and evaluate the combat performance of the UAV according to the combat performance evaluation value of the UAV.

[0085] Among them, the calculation formula for determining the combat performance evaluation value of the UAV based on the strike benefit data set and the matching rate is:

[0086]

[0087] Wherein, E is the combat performance evaluation value of the UAV; (P1, P2,..., P d ) is the strike benefit data set; d is the number of UAVs; k1 is the weight factor of the strike benefit, and the value range is from 0.40 to 0.65; k2 is the weight factor of the matching rate, and the value range is from 0.35 to 0.60; M is the matching rate, which is the ratio of the predicted matching number to the actual matching number.

[0088] Compare the UAV combat performance evaluation method of the present invention with the existing comprehensive algorithm. Under the conditions of given prior probability, state transition probability and observed variable values at different times, substitute the weighted dynamic Bayesian algorithm to calculate the probabilities that the strike benefit levels at different times belong to each state. The specific results are shown in Table 3. The weights of the evaluation indexes are determined by the grey relational algorithm as (0.153, 0.136, 0.121, 0.148, 0.161, 0.124, 0.057, 0.1).

[0089] By comparing Table 2 and Table 3, it can be seen that at time T1, multiple evaluation index parameters of the UAV's perception ability and tracking ability are at relatively low levels. Although the real-time performance has certain advantages, the strike effectiveness level is still judged to be low. At time T3, the relative speed of the UAV has increased, and the parameters of the perception ability and tracking ability have also reached a medium level. However, the real-time performance has decreased. At this time, the strike effectiveness level is judged to be medium. At time T4, the parameters of the UAV's perception ability and tracking ability are both at a high level, and the real-time performance is also at a medium level. The strike effectiveness level is judged to be high.

[0090] Table 2: Observation node variable data at different times

[0091]

[0092] Table 3: Probability of strike effectiveness level at different times

[0093] Strike effectiveness level T1 T2 T3 T4 High 0.102 0.102 0.412 0.768 Medium 0.276 0.276 0.495 0.159 Low 0.622 0.622 0.093 0.073

[0094] See Figure 4 , the strike effectiveness probabilities obtained at time T1 and time T2 are similar. Since the observed variable values at time T2 and time T1 show similarity after the similarity measure calculation, therefore, the strike effectiveness level probabilities of the UAV combat performance evaluation method of the present invention are the same at the two times.

[0095] See Figure 2 , the present invention also provides a UAV combat performance evaluation system, which is applied to the above-mentioned UAV combat performance evaluation method, and includes a Bayesian network construction module, an improvement module, a strike effectiveness value determination module, a strike effectiveness data set determination module, a matching module, and a combat performance evaluation module;

[0096] The Bayesian network construction module is used to determine multiple evaluation indexes of the UAV combat performance, and construct a Bayesian network according to the correlation relationship between the multiple evaluation indexes;

[0097] The improvement module: is used to improve the Bayesian network by using the grey relational algorithm to form a weighted Bayesian algorithm;

[0098] The strike effectiveness value determination module: is used to determine the strike effectiveness value of the UAV against the target at the current moment by using the Euclidean distance metric algorithm and the weighted Bayesian algorithm; is used to construct a weighted dynamic Bayesian algorithm based on the weighted Bayesian algorithm, and determine the strike effectiveness value between the UAV and its matched target according to the strike effectiveness value of the UAV against the target at the current moment and the weighted dynamic Bayesian algorithm;

[0099] The strike effectiveness data set determination module: is used to repeat the improvement module and the strike effectiveness value determination module until the strike effectiveness values between all UAVs and their matched targets are determined to form a strike effectiveness data set;

[0100] Matching module: used to determine the matching rate between the UAV and the target;

[0101] And the combat performance evaluation module: used to determine the combat performance evaluation value of the UAV based on the strike benefit data set and the matching rate, and evaluate the combat performance of the UAV according to the combat performance evaluation value of the UAV.

[0102] The UAV combat performance evaluation system of the present invention corresponds to the above UAV combat performance evaluation method. Among them, for the specific contents of the Bayesian network construction module, improvement module, strike benefit value determination module, strike benefit data set determination module, matching module and combat performance evaluation module, refer to the description in the above UAV combat performance evaluation method part, and the present invention will not elaborate here.

[0103] The present invention also provides a memory, on which a program file is stored, and the program file is executed to implement the program instructions formed by the above UAV combat performance evaluation method.

[0104] The memory in the present invention may specifically include a random access memory (RAM), internal memory, read-only memory (ROM), programmable ROM, erasable programmable ROM, register, hard disk, removable disk or CD-ROM. It should be noted that those skilled in the art can specifically select the form and type of the storage medium according to actual usage requirements, and the present invention does not make further specific limitations.

[0105] The present invention also provides an electronic device, including a processor and a memory that are coupled to each other. Among them, the memory: used to store the program instructions formed by the above UAV combat performance evaluation method; the processor: used to execute the program instructions stored in the memory.

[0106] The electronic device in the present invention includes any electronic device that can execute program instructions, such as a computer, a mobile terminal, a remote control device or a wearable device, etc.

[0107] The above content is only used to illustrate the technical solution of the present invention, rather than a limitation to the present invention; although the present invention has been described in detail with reference to the foregoing, those of ordinary skill in the art should understand that: they can still modify the foregoing described technical solution, or perform equivalent replacement on some or all of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solution deviate from the scope of the technical solution of the present invention.

Claims

1. A method for evaluating the combat performance of drones, characterized in that, Including the following steps: Determine multiple evaluation indicators of the combat performance of the UAV, and construct a Bayesian network according to the correlation relationship between the multiple evaluation indicators; Use the grey relational algorithm to improve the Bayesian network to form a weighted Bayesian algorithm; Use the Euclidean distance metric algorithm and the weighted Bayesian algorithm to determine the strike effectiveness value of the UAV against the target at the current moment; construct a weighted dynamic Bayesian algorithm based on the weighted Bayesian algorithm, and determine the strike effectiveness value between the UAV and its matching target according to the strike effectiveness value of the UAV against the target at the current moment and the weighted dynamic Bayesian algorithm; wherein, the specific process of using the Euclidean distance metric algorithm and the weighted Bayesian algorithm to determine the strike effectiveness value of the UAV against the target at the current moment includes: use the Euclidean distance metric algorithm to judge the similarity of the evaluation indicators corresponding to two adjacent moments, if the similarity is greater than the similarity threshold, then use the strike effectiveness value of the previous moment as the strike effectiveness value of the UAV against the target at the current moment; if the similarity does not exceed the similarity threshold, then determine the strike effectiveness value of the UAV against the target at the current moment according to the weighted Bayesian algorithm; Repeat the above steps until the strike effectiveness values between all UAVs and their matching targets are determined to form a strike effectiveness data set; Determine the matching rate between the UAV and the target; wherein, the specific process of determining the matching rate between the UAV and the target is: determine the threat level of the target, and match the UAV and the target according to the threat level of the target and the strike effectiveness value between the UAV and its matching target, and determine the matching rate; Determine the combat performance evaluation value of the UAV based on the strike effectiveness data set and the matching rate, and evaluate the combat performance of the UAV according to the combat performance evaluation value of the UAV; Wherein, the calculation formula for determining the combat performance evaluation value of the UAV based on the strike effectiveness data set and the matching rate is: Where E is the combat performance evaluation value of the UAV; (P1, P2, …, P d ) is the strike benefit data set; d is the number of UAVs; k1 is the weight factor of the strike benefit, and its value range is from 0.40 to 0.65; k2 is the weight factor of the matching rate, and its value range is from 0.35 to 0.60; M is the matching rate, which is the ratio of the predicted matching quantity to the actual matching quantity.

2. The method for evaluating the combat performance of an unmanned aerial vehicle according to claim 1, wherein The specific process of using the grey relational algorithm to improve the Bayesian network to form a weighted Bayesian algorithm includes: Use the grey relational algorithm to assign different weights to each node of the Bayesian network, and determine the weights of each sub-node in the Bayesian network relative to the root node in combination with the different weights of each node, so as to form a weighted Bayesian algorithm.

3. The method for evaluating the combat performance of an unmanned aerial vehicle according to claim 1, wherein, The specific process of constructing a weighted dynamic Bayesian algorithm based on the weighted Bayesian algorithm includes: Add time series data to the weighted Bayesian algorithm to construct a weighted dynamic Bayesian algorithm.

4. The method for evaluating the combat performance of an unmanned aerial vehicle according to claim 1, wherein, The evaluation indicators include strike effectiveness, perception ability, tracking ability, maneuverability and real-time performance; wherein, the perception ability includes false alarm rate, miss detection rate and positioning accuracy, and the maneuverability includes relative speed, relative altitude and heading angle.

5. Drone combat performance evaluation system, applied to the drone combat performance evaluation method described in claim 1, characterized in that, Including: A Bayesian network construction module, used to determine multiple evaluation indicators of the combat performance of the UAV, and construct a Bayesian network according to the correlation relationship between the multiple evaluation indicators; An improvement module: used to use the grey relational algorithm to improve the Bayesian network to form a weighted Bayesian algorithm; Strike effectiveness value determination module: used to determine the strike effectiveness value of the UAV against the target at the current moment by using the Euclidean distance metric algorithm and the weighted Bayesian algorithm; used to construct a weighted dynamic Bayesian algorithm based on the weighted Bayesian algorithm, and determine the strike effectiveness value between the UAV and its matching target according to the strike effectiveness value of the UAV against the target at the current moment and the weighted dynamic Bayesian algorithm; wherein, the specific process of determining the strike effectiveness value of the UAV against the target at the current moment by using the Euclidean distance metric algorithm and the weighted Bayesian algorithm includes: using the Euclidean distance metric algorithm to judge the similarity of the evaluation indicators corresponding to two adjacent moments, if the similarity is greater than the similarity threshold, then use the strike effectiveness value of the previous moment as the strike effectiveness value of the UAV against the target at the current moment; if the similarity does not exceed the similarity threshold, then determine the strike effectiveness value of the UAV against the target at the current moment according to the weighted Bayesian algorithm. Strike effectiveness data set determination module: used to repeat the improvement module and the strike effectiveness value determination module until the strike effectiveness values between all UAVs and their matching targets are determined, forming a strike effectiveness data set. Matching module: used to determine the matching rate between the UAV and the target; wherein, the specific process of determining the matching rate between the UAV and the target is: determine the threat level of the target, and match the UAV and the target according to the threat level of the target and the strike effectiveness value between the UAV and its matching target, and determine the matching rate. And the combat performance evaluation module: used to determine the combat performance evaluation value of the UAV based on the strike effectiveness data set and the matching rate, and evaluate the combat performance of the UAV according to the combat performance evaluation value of the UAV. Wherein, the calculation formula for determining the combat performance evaluation value of the UAV based on the strike effectiveness data set and the matching rate is: where E is the combat performance evaluation value of the UAV; (P1, P2, …, P d ) is the strike benefit data set; d is the number of UAVs; k1 is the weight factor of the strike benefit, and its value range is from 0.40 to 0.65; k2 is the weight factor of the matching rate, and its value range is from 0.35 to 0.60; M is the matching rate, which is the ratio of the predicted matching quantity to the actual matching quantity.

6. A memory, characterized in that, There is a program file stored, and the program file is executed to implement the program instructions formed by the UAV combat performance evaluation method according to any one of claims 1-4.

7. An electronic device, characterized in that, It includes a processor and a memory that are coupled to each other, wherein, The memory: used to store the program instructions formed by the UAV combat performance evaluation method according to any one of claims 1-4; The processor: used to execute the program instructions stored in the memory.

Images