Cooperative detection and task allocation method and device under anti-unmanned aerial vehicle air defense system

Abstract

The present application belongs to the field of command and control, and relates to a cooperative detection and task allocation method and device under an anti-UAV air defense system, which cooperatively accesses radar, photoelectric and radio frequency detection and interception equipment, realizes heterogeneous device interconnection and data standardization. Through plot-track association fusion, the same target is identified and the track is interrupted, and a unified air situation map is formed. The target motion, intention and physical properties are extracted, the multi-dimensional threat factors are quantified and the near defense time domain is corrected. According to the threat evaluation, combined with the interception success rate, cost and availability, a multi-constraint optimization model is constructed to complete the optimal assignment of detection and interception tasks. In execution, the situation is monitored in real time, the uncompleted tasks are quickly redistributed when changes occur, and the strategy is corrected through effect evaluation, forming a closed loop of detection-evaluation-decision-execution-feedback. The present application adopts a systematic cooperative architecture, changes single equipment into networked intelligence warfare, and has the capabilities of multi-source track association fusion, multi-constraint task allocation and dynamic re-planning.

Description

Technical Field

[0001] This invention relates to the field of command and control technology, and in particular to a method and apparatus for collaborative detection and task allocation in an anti-drone air defense system. Background Technology

[0002] Currently, the threat posed by drones exhibits a trend towards low-altitude, slow-speed, miniaturized (low, slow, and small), swarm-based, and intelligent development. Traditional anti-drone equipment is mostly deployed independently, operating in silos, resulting in severe information isolation and insufficient continuous tracking and relay interception capabilities against drone targets, especially small low-altitude penetrating targets, making it difficult to form a systematic defense force. This problem is particularly prominent in radar blind zones (below 150 meters)—once a target is lost, it is difficult to re-acquire it, leading to missed interception windows. Summary of the Invention

[0003] To address the aforementioned technical problems, this invention provides a collaborative detection and task allocation method within an anti-drone air defense system, employing the following technical solution, including the following steps: The platform integrates the dispersed radar detection nodes, photoelectric detection nodes, radio frequency detection nodes, and various soft and hard kill interception equipment within the region into a unified collaborative command and control platform, establishing a unified time reference and spatial coordinate system to achieve interconnection and interoperability of heterogeneous equipment and standardized data aggregation. The system performs point-track-track association on the aggregated multi-source heterogeneous detection data, identifies heterogeneous sensor measurements belonging to the same target, re-associates interrupted track segments within radar blind zones, and generates a unified air situation map through data fusion. Based on the generated unified air situation map, the motion characteristics, attack intentions and physical attributes of each tracked target are extracted, multi-dimensional threat factors are quantified, the threat index of each target is calculated in a comprehensive manner, and the threat time domain correction is performed on targets that continue to approach the protected target. Based on the threat assessment results, the success probability of each interception device against different targets is estimated, the interception cost of each interception device is quantified, and the current occupancy status and resource availability of each device are perceived in real time to determine the availability of the devices. Based on the target threat index and the effectiveness, cost and availability parameters of interception equipment, a multi-constraint multi-target task allocation optimization model is established to solve the optimal assignment scheme of detection and interception tasks and generate the execution instruction sequence of each piece of equipment. During mission execution, the system monitors situational changes in real time. When triggering conditions are met, it initiates rapid replanning, reassigns unfinished interception tasks, and corrects subsequent strategies through interception effectiveness evaluation, forming a closed loop of detection-evaluation-decision-execution-feedback.

[0004] To address the aforementioned technical problems, this invention also provides a collaborative detection and task allocation device within an anti-drone air defense system, employing the following technical solution, including: A module is established to unify the access of dispersed radar detection nodes, photoelectric detection nodes, radio frequency detection nodes, and various soft and hard kill interception equipment to the collaborative command and control platform, establish a unified time reference and spatial coordinate system, and realize the interconnection and interoperability of heterogeneous equipment and the standardized aggregation of data. The identification module is used to perform point-track association on the aggregated multi-source heterogeneous detection data, identify heterogeneous sensor measurements belonging to the same target, re-associate interrupted track segments in the radar blind zone, and generate a unified air situation map through data fusion. The extraction module is used to extract the motion characteristics, attack intentions and physical attributes of each tracked target based on the generated unified air situation map, quantify multi-dimensional threat factors, comprehensively calculate the threat index of each target, and perform threat time-domain correction for targets that are continuously approaching the protected target. The quantification module is used to estimate the success probability of each interception device against different targets based on the threat assessment results, quantify the interception cost of each interception device, and perceive the current occupancy status and resource availability of each device in real time to determine the availability of the devices. A module is established to create a multi-constraint, multi-target task allocation optimization model based on the target threat index and the effectiveness, cost, and availability parameters of interception equipment. This model solves for the optimal assignment scheme of detection and interception tasks and generates the execution command sequence for each piece of equipment. The correction module is used to monitor situational changes in real time during task execution. When the triggering conditions are met, it initiates rapid replanning, reassigns unfinished interception tasks, and corrects subsequent strategies based on the interception effect evaluation, forming a closed loop of detection-evaluation-decision-execution-feedback.

[0005] To address the aforementioned technical problems, the present invention also provides a computer device that employs the technical solution described below, comprising a memory and a processor. The memory stores computer-readable instructions, and the processor executes the computer-readable instructions to implement the steps of the aforementioned collaborative detection and task allocation method under the anti-drone air defense system.

[0006] To address the aforementioned technical problems, the present invention also provides a computer-readable storage medium, which employs the technical solution described below. The computer-readable storage medium stores computer-readable instructions, which, when executed by a processor, implement the steps of the aforementioned collaborative detection and task allocation method under the anti-drone air defense system.

[0007] Compared with the prior art, the present invention has the following main advantages: (1) Adopting a systematic collaborative architecture to upgrade single-equipment independent combat to network-based intelligent combat: Through networking and the establishment of a unified benchmark, the dispersed radar, photoelectric, radio frequency detection nodes and various interception equipment are transformed from information islands into resource pools that can be collaboratively scheduled, realizing the sharing of detection resources and the relay of interception tasks.

[0008] (2) Multi-source track association and fusion: Through hierarchical clustering and Hausdorf distance reassociation technology, the problem of difficulty in re-acquiring targets lost in the radar blind zone (below 150 meters) has been solved, and continuous tracking and maintenance under multi-node relay detection has been achieved, which significantly expands the effective coverage of the defense system.

[0009] (3) Multi-constraint task allocation: Through threat assessment, effectiveness cost modeling, and multi-objective optimization, low-cost interference means are prioritized to deal with low-threat targets, while high-cost hard-kill means are reserved to deal with high-threat targets, thus optimizing the overall defense cost-effectiveness ratio.

[0010] (4) Dynamic replanning closed-loop optimization: Through real-time situation detection and rolling optimization, the system can adapt to dynamic changes in the situation (swarm attack, simultaneous attack from multiple directions, target maneuvering and evasion), which improves the success rate of defense against sudden scenarios. Attached Figure Description

[0011] To more clearly illustrate the solutions in this invention, the accompanying drawings used in the description of the embodiments of this invention will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0012] Figure 1 This is a flowchart of an embodiment of the cooperative detection and task allocation method under the anti-drone air defense system of the present invention; Figure 2 This is a schematic diagram of the structure of an embodiment of the collaborative detection and task allocation device in the anti-drone air defense system of the present invention; Figure 3 This is a schematic diagram of the structure of one embodiment of the computer device of the present invention. Detailed Implementation

[0013] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains; the terminology used herein in the specification is for the purpose of describing particular embodiments only and is not intended to limit the invention; the terms "comprising" and "having," and any variations thereof, in the specification, claims, and foregoing drawings are intended to cover non-exclusive inclusion. The terms "first," "second," etc., in the specification, claims, or foregoing drawings are used to distinguish different objects and not to describe a particular order.

[0014] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of the invention. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0015] To enable those skilled in the art to better understand the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings.

[0016] It should be noted that the collaborative detection and task allocation method under the anti-drone air defense system provided in the embodiments of the present invention is generally executed by a server / terminal device, and correspondingly, the collaborative detection and task allocation device under the anti-drone air defense system is generally set in the server / terminal device.

[0017] It should be understood that the number of terminal devices, networks, and servers is merely illustrative. Depending on implementation needs, any number of terminal devices, networks, and servers can be used. Example

[0018] Please refer to Figure 1 The flowchart illustrates an embodiment of the cooperative detection and task allocation method under the anti-drone air defense system of the present invention. The cooperative detection and task allocation method under the anti-drone air defense system includes the following steps: Step S1 involves integrating the dispersed radar detection nodes, photoelectric detection nodes, radio frequency detection nodes, and various soft and hard kill interception equipment within the region into the collaborative command and control platform, establishing a unified time reference and spatial coordinate system, and realizing the interconnection and interoperability of heterogeneous equipment and the standardized aggregation of data.

[0019] In this embodiment, the electronic equipment (e.g., server / terminal equipment) running on the cooperative detection and task allocation method within the anti-drone air defense system can receive cooperative detection and task allocation requests from the anti-drone air defense system via wired or wireless connections. It should be noted that the aforementioned wireless connection methods may include, but are not limited to, 3G / 4G / 5G connections, WiFi connections, Bluetooth connections, WiMAX connections, Zigbee connections, UWB (ultra-wideband) connections, and other currently known or future-developed wireless connection methods.

[0020] In this embodiment, step S1 may specifically include the following steps: S11 uses a layered access adaptation architecture to configure corresponding physical access modules for the communication interfaces of different equipment. Through protocol conversion middleware, the private communication protocols of each piece of equipment are uniformly converted into standard internal data exchange protocols. Each equipment node registers to the collaborative command and control platform through a standardized agent, and the platform maintains the global equipment registry.

[0021] A layered access adaptation architecture is designed. At the physical layer, corresponding physical access modules are configured for the communication interfaces of different equipment (wired Ethernet, wireless data link, fiber optic, etc.). At the protocol layer, a protocol conversion middleware is built to uniformly convert the private protocols of each piece of equipment (such as the customized TCP / IP protocol for radar, the serial port protocol for optoelectronic equipment, and the CAN bus protocol for interception equipment) into a standard internal data exchange protocol. The adaptation middleware adopts a publish-subscribe model. Each equipment node registers with the collaborative command and control platform through a standardized agent. The platform maintains a global equipment registry, recording the type, location, capability parameters, IP address, and current online status of each node. The registration information is encapsulated in JSON format, including fields such as unique equipment identifier, latitude and longitude coordinates, detection / interception capability description, maximum effective range, and current status.

[0022] The purpose of step S11 is to resolve the connectivity issues caused by heterogeneous data interfaces, communication protocols, and data formats of various types of equipment, and to achieve standardized aggregation of detection data and equipment status information.

[0023] S12 uses the BeiDou satellite timing module as a unified time source, deploys BeiDou timing receivers at each access node, performs time synchronization calibration in the local area network through a precise time protocol, and uses a recursive least squares algorithm to dynamically compensate for clock drift, so that all nodes in the network maintain a unified time reference, and each data packet after synchronization is accompanied by a globally unified timestamp.

[0024] The BeiDou satellite time synchronization module is used as the unified time source. A BeiDou time receiver is deployed at each access node to receive the standard time signal broadcast by the BeiDou satellite and perform time synchronization calibration within the local area network using the IEEE 1588 Precise Time Protocol (PTP). The specific process is as follows: (1) The Beidou receiver acquires UTC time and uses it as the master clock; (2) Each node exchanges synchronization messages with the master clock via the PTP protocol and calculates the link delay and clock offset; (3) The Recursive Least Squares (RLS) algorithm is used to dynamically compensate for clock drift, controlling the time synchronization accuracy of all network nodes to the microsecond level. Each synchronized data packet is accompanied by a globally unified timestamp t. global The unit is seconds (s), and the format is t. global =t GPS +Δt comp , where t GPS The pulse time per second calculated by the BeiDou receiver, in seconds (s), Δt comp The cumulative offset correction of the local clock relative to the master clock, calculated for the PTP protocol, in seconds (s).

[0025] The purpose of step S12 is to eliminate timing inconsistencies caused by clock drift at each node, ensure the comparability of multi-source data in the time dimension, and provide an accurate timestamp reference for track association.

[0026] S13. Select the geocentric rectangular coordinate system or the station-centered coordinate system as the common reference coordinate system, transform the measurement data of each detection node in different local coordinate systems to the common reference coordinate system, and use the error propagation analysis method to estimate the measurement error after coordinate transformation.

[0027] Choose either the geocentric rectangular coordinate system (ECEF) or the station-centered coordinate system (ENU) as the common reference coordinate system. The coordinate transformation process is as follows: (1) Conversion of local coordinates of each node to WGS-84 geodetic coordinates.

[0028] For node i, its measurement in the local coordinate system is: The known WGS-84 geodetic coordinates of the node are (B i ,L i H i (Latitude, Longitude, Altitude), first convert to ECEF coordinates. : .

[0029] in, , , , — The three-dimensional coordinates of node i in the geocentric rectangular coordinate system (ECEF), in meters (m).

[0030] B i — The geodetic latitude of node i, in radians (rad).

[0031] L i — The geodetic longitude of node i, in radians (rad).

[0032] H i — The ground elevation of node i, in meters (m).

[0033] N i — Radius of curvature of the circle (Maoyou Circle) The unit is meters (m). 2 =0.00669437999014 is the square of the first eccentricity.

[0034] a — semi-major axis of the WGS-84 ellipsoid, a=6378137 m.

[0035] (2) Propagation of mean and covariance of target coordinate transformation error at measurement points.

[0036] Since the transformation involves nonlinear functions, a first-order Taylor expansion (unscented transformation UT) is used to perform propagation analysis on the measurement error, obtaining the mean error μ after the transformation. e The covariance matrix P e : μ e ≈ ∂f / ∂x μ x , P e ≈ ∂f / ∂x P x (∂f / ∂x) T +∂f / ∂θ P θ (∂f / ∂θ) T .

[0037] Where: μ e — The mean vector of the error after coordinate transformation.

[0038] P e — The covariance matrix of the error after coordinate transformation.

[0039] f(⋅) — Coordinate transformation function that maps measurements in the local coordinate system to a common reference coordinate system.

[0040] x — Target measurement vector (such as distance, azimuth, elevation, etc.).

[0041] θ — The sensor node's own position parameter vector (such as latitude, longitude, and altitude).

[0042] μ x — The mean vector of the measurement noise.

[0043] P x — Measurement noise covariance matrix.

[0044] P θ — Node position error covariance matrix.

[0045] ∂f / ∂x — Jacobian matrix of the coordinate transformation function with respect to the measurement vector x.

[0046] ∂f / ∂θ — Jacobian matrix of the coordinate transformation function with respect to the node position parameter vector θ.

[0047] (⋅) T — Matrix transpose operation.

[0048] This error propagation model provides a probabilistic error boundary for the design of association thresholds in subsequent track association.

[0049] After the above transformation, the measurement data of all detection nodes are unified under a common spatiotemporal reference, providing data input for multi-source track association in step S2.

[0050] The purpose of step S13 is to uniformly transform the measurement data of each node in different coordinate systems (such as the spherical coordinate system of radar, the pixel coordinate system of photoelectric, and the WGS-84 latitude and longitude coordinate system of GPS) to a common reference coordinate system, thereby eliminating errors caused by inconsistent spatial references.

[0051] The purpose of step S1 is to construct the physical and information infrastructure for anti-drone collaborative processing, to integrate the radar, optoelectronic, and radio frequency detection nodes and various interception equipment that are scattered in the area into the collaborative command and control platform, to establish a unified time reference and spatial coordinate system, and to provide standardized input for subsequent multi-source data fusion and task allocation.

[0052] Step S2 involves performing point-track-track association on the aggregated multi-source heterogeneous detection data, identifying heterogeneous sensor measurements belonging to the same target, re-associating interrupted track segments within the radar blind zone, and generating a unified air situation map through data fusion.

[0053] In this embodiment, step S2 may specifically include the following steps: S21. For the established target trajectory, the predicted state and predicted covariance of the trajectory are obtained through a filtering algorithm. The normalized information distance between the newly arrived detection point and each trajectory is calculated. An elliptical correlation gate is used for correlation decision. A stop-and-go model is added to the motion model to adapt to the hovering and low-speed maneuvering characteristics of low, slow and small targets.

[0054] To address the characteristics of low-speed, small targets, such as slow flight speeds and the potential for hovering or low-speed maneuvering, a nearest-neighbor association algorithm with an added stop-and-go motion model is employed. This algorithm is used for predicting the trajectory state. (k|k-1) and prediction covariance P j (k|k-1), calculate the normalized innovation distance (Mahanobis distance) between the new point z(k) and the track. ij : .

[0055] in: — The squared Mahalanobis distance (dimensionless) between point i and track j.

[0056] z i (k) — Measurement vector (e.g., position coordinates) obtained by the i-th sensor at time k.

[0057] H — Measurement matrix, which maps the state vector to the measurement space.

[0058] (k|k-1) — The predicted state vector of track j at time k.

[0059] S j (k) — the new covariance matrix of track j, S j (k)=HP j (k|k-1) H T +R j R j To measure the noise covariance.

[0060] (⋅) -1 — Matrix inversion operation.

[0061] (⋅) T — Matrix transpose operation.

[0062] The association decision uses an elliptic association gate strategy: if If ≤γ (γ is determined by looking up a table using the chi-square distribution, usually taking the threshold value corresponding to the confidence interval), then the point i falls within the correlation gate of the track j. To address the stopping-and-go characteristics of low-speed, small targets, a Markov switching probability from the Stop-and-Go Model is added to the motion model. This ensures that the filter maintains its estimation of the target position even when the target velocity approaches zero, avoiding misclassification of hovering targets as false points or loss of tracking.

[0063] The purpose of step S21 is to coarsely correlate the newly arrived sensor detection points with the existing target tracks to preliminarily determine whether the points belong to the tracked targets, in order to deal with the characteristics of low, slow and small targets that are slow and may stop moving.

[0064] S22. Extract the motion feature vectors of each track segment before and after the interruption. Use the Hausdorff distance between track segments as a similarity measure. Employ a bottom-up agglomerative hierarchical clustering method to reassociate interrupted track segments belonging to the same target. Simultaneously constrain the consistency of motion features during the clustering process.

[0065] A track association method combining hierarchical clustering and Hausdorff distance is adopted.

[0066] Perform feature extraction for each track segment T: a Extract its motion feature vector: .

[0067] Wherein: F a For track segment T a The motion feature vector, This is the average velocity vector (including magnitude and direction) of the trajectory segment. p is the average acceleration vector of this trajectory segment. start and p end These are the start and end positions of the trajectory segment, Δt. a The duration of the track is expressed in seconds (s).

[0068] Perform Hausdorff distance calculation: for two track segments T a and T b Its Hausdorf is far from d H (T a ,T b ) is defined as: .

[0069] Where: d H (T a ,T b — Flight track fragment T a With T b The Hausdorff distance between them, in meters (m).

[0070] sup — Supremum operation (takes the least upper bound of the set).

[0071] inf — the infimum operator (takes the largest lower bound in the set).

[0072] p — track fragment T a A vector of any point in the vector.

[0073] q — Flight track fragment T b A vector of any point in the vector.

[0074] ||pq|| — The Euclidean distance between points p and q, in meters (m).

[0075] max{⋅, ⋅} — Takes the larger of the two values.

[0076] Hausdorff distance measures the maximum and minimum distance between two sets of track points and has a good ability to characterize the similarity of track shapes.

[0077] Hierarchical clustering association is performed: Hausdorff distance is used as the similarity measure between track segments, and a bottom-up agglomerative hierarchical clustering is adopted: initially, each track segment forms its own cluster, and then the two closest clusters are gradually merged until the minimum distance between clusters exceeds a preset threshold θ. H During the clustering process, the consistency constraint of motion characteristics is considered (the change in velocity and acceleration direction does not exceed 30°) to avoid misassociating the tracks of different targets.

[0078] This method can effectively address situations where low, slow, and small targets reappear after being lost in radar blind zones, resulting in sparse track points and potential changes in motion patterns. It uses multi-node detection data to form track segments of sufficient density for matching and reassociation.

[0079] The purpose of step S22 is to solve the problem of difficulty in re-acquiring targets lost in the radar blind zone (below 150 meters). When a target crosses the radar blind zone or is affected by low-altitude clutter, causing the track to be interrupted, the track segments before and after the interruption are re-linked into the same target trajectory through multi-node relay detection.

[0080] S23, the measurements from multiple sensors associated with the same target are fused, and the inverse matrix of the covariance matrix of each sensor's state estimate is used as the information weight. The state estimates of each sensor are weighted and summed to obtain the fused target state and its covariance, and a unified air situation map is generated.

[0081] A covariance-weighted fusion algorithm (also known as simple convex combination fusion) is employed. Let target j have N at time k. j (k) sensors detect the target simultaneously, and each sensor estimates the target's state as follows: The corresponding estimated covariance is The target state after fusion Covariance for: ; .

[0082] in: — The state covariance matrix of target j at time k after fusion.

[0083] — The covariance matrix of the state estimation of target j by the i-th sensor at time k.

[0084] N j (k) — The number of sensors that can detect target j at time k.

[0085] ∑ — Summation operation.

[0086] (⋅) -1 — Matrix inversion operation.

[0087] The estimates from each sensor are weighted according to the information matrix (inverse covariance matrix). Sensors with smaller covariance (more accurate estimates) are given greater weight, and the fused covariance is better than the estimation accuracy of any single sensor. For heterogeneous sensors (such as radar and photoelectric sensors), the measurements must first be unified to a common coordinate system through coordinate transformation before fusion.

[0088] After this step, the collaborative command and control platform generates a unified air situation map that includes the position, speed, heading, RCS characteristics, and target type (classification) of all targets.

[0089] The purpose of step S23 is to fuse all sensor measurements associated with the same target into an accurate target state estimate, generate a unified air situation map, and provide accurate target position, speed, and heading information for target threat assessment and task allocation.

[0090] The purpose of step S2 is to correlate and fuse the multi-source heterogeneous detection data (radar spot data, photoelectric image target coordinates, radio frequency detection signal characteristics, etc.) collected in step S1, identify the sensor measurements belonging to the same target, generate a unified air situation map, and solve the problem of multi-source data being aligned with the same target.

[0091] Step S3: Based on the generated unified air situation map, extract the motion characteristics, attack intentions and physical attributes of each tracked target, quantify multi-dimensional threat factors, comprehensively calculate the threat index of each target, and perform threat time-domain correction on targets that are continuously approaching the protected target.

[0092] In this embodiment, step S3 may specifically include the following steps: S31 extracts normalized threat factors from five dimensions: target speed, target heading towards the degree of protection of the target, target flight altitude, target radar cross-section, and number of targets.

[0093] Define five threat factors, each normalized to the [0,1] interval: Target speed threat factor T v :T v=min(1,(vv min ) / (v max -v min The higher the speed, the shorter the reaction time, and the greater the threat.

[0094] Wherein: T v — Target speed threat factor (dimensionless, value range [0,1]).

[0095] v — the target's current velocity, in meters per second (m / s).

[0096] v min — Threat speed lower limit, such as 20 m / s.

[0097] v max — Threat speed limit, such as 100 m / s.

[0098] min(⋅) — Minimum value operation, used to truncate factors to within 1.

[0099] Target heading threat factor T c : .

[0100] Wherein: T c — Target heading threat factor (dimensionless, value range (0,1]).

[0101] θ LOS — The angle between the target's heading and the line connecting the target and the protected target, expressed in radians (rad).

[0102] σ θ — Angle scale parameter, usually taken as 15° (equivalent to approximately 0.2618 radians).

[0103] exp(⋅) — the natural exponential function.

[0104] This formula is based on the Gaussian kernel function, and T is used when the heading is directly facing the protected target. c →1, deviation T c Rapid decay.

[0105] The degree to which the course is directed toward important protected targets (such as command posts or airports).

[0106] Target height threat factor T h :T h =exp(-h / h0).

[0107] Wherein: T h — Target height threat factor (dimensionless, value range (0,1]).

[0108] h — Current height of the target, in meters (m).

[0109] h0 — Dimensional height, usually taken as 100 m.

[0110] Low-altitude penetrating targets pose a greater threat than high-altitude targets (because low-altitude targets are more difficult to detect and intercept).

[0111] Target RCS characteristic threat factor T rcs :T rcs =max(0,1-RCS / (RCS0)).

[0112] Wherein: T rcs — Target RCS characteristic threat factor (dimensionless, value range [0,1]).

[0113] RCS — Radar Cross Section of the Target, in square meters (m²).

[0114] RCS0 — Reference cross-sectional area, usually taken as 1 m² 2 .

[0115] max(⋅) — The maximum value operation, used to truncate factors below 0.

[0116] The smaller the RCS, the more difficult the target is to detect, the stronger its concealment, and the greater its relative threat (but this needs to be judged in conjunction with other factors).

[0117] Target quantity threat factor T n :T n =min(1,N target / N sat ).

[0118] Wherein: T n — Target quantity threat factor (dimensionless, value range [0,1]). N target —Total number of targets in the current area. N sat — The maximum number of targets that the defense system can handle simultaneously.

[0119] In swarm attack scenarios, the sheer number of targets poses a saturation attack threat.

[0120] The purpose of step S31 is to extract threat quantification indicators from three dimensions: target motion characteristics, attack intent, and physical attributes.

[0121] S32: The five-dimensional threat factors are used to form a factor vector. The weight vector of each factor is determined by the analytic hierarchy process. The factors are mapped to a preset evaluation set through the membership function. A fuzzy relation matrix is ​​constructed and fuzzy comprehensive synthesis is performed to obtain the comprehensive threat index of each target.

[0122] A fuzzy comprehensive evaluation method is adopted. Let the threat factor vector T = [T v ,T c ,T h ,T rcs ,T n Let the evaluation set V = {v1, v2, v3, v4} correspond to low threat, medium threat, high threat, and extremely high threat, respectively. A fuzzy relation matrix R is constructed by mapping each factor to the evaluation set using a membership function. The factor weight vector W = [w1, w2, w3, w4, w5] is determined using the Analytic Hierarchy Process (AHP). The fuzzy comprehensive evaluation result is: B = W∘R.

[0123] Where: B — Fuzzy comprehensive evaluation result vector, containing 4 elements, corresponding to the membership degree of each evaluation level.

[0124] W — Factor weight vector (a 1×5 row vector), where each element wi satisfies ∑ i=1 5 w i =1.

[0125] R — Fuzzy relation matrix (a 5×4 matrix), element r ij This represents the membership degree of the i-th factor to the j-th evaluation level.

[0126] ∘ — Fuzzy synthesis operator, using weighted average type operation, using weighted average type: b j =∑ i=1 5 w i ⋅ r ij , j=1,2,3,4. Final Threat Index TI=∑ j=1 4 b j ⋅μ j .

[0127] Where: TI — Target Comprehensive Threat Index (dimensionless, value range [0,1]), the higher the value, the higher the threat.

[0128] b j — The degree of membership of the evaluation result to the j-th evaluation level.

[0129] μ j — Rating level v j The corresponding center value is μ = [0.2, 0.5, 0.8, 1.0].

[0130] The purpose of step S32 is to integrate multidimensional factors into a single threat index, avoid misjudgment by a single factor, and use fuzzy logic to handle the complex nonlinear relationships between factors.

[0131] S33, calculate the expected arrival time of each target to the protected target, construct a time-domain correction function based on the expected arrival time, and dynamically correct the comprehensive threat index so that the threat index of targets that are continuously approaching the protected target increases over time.

[0132] Calculate the estimated time of arrival (ETA) for each target: ETA = D protect / v radial .

[0133] Where: ETA — the estimated time for the target to reach the protected target, in seconds (s).

[0134] D protect — The straight-line distance between the target and the protected target, in meters (m).

[0135] v radial — Target radial velocity (the velocity component in the direction of the target position vector, with positive values ​​close to), in meters per second (m / s).

[0136] Threat Index revised to: TI 修正 (t)=TI0⋅[1+α⋅exp(-ETA / τ)].

[0137] Among them: TI 修正 (t) — Threat index (dimensionless) after correction at time t.

[0138] TI0 — The baseline threat index (dimensionless) obtained from fuzzy comprehensive evaluation at the current moment.

[0139] α — Corrected intensity coefficient, usually taken as 0.3 (dimensionless).

[0140] τ — Time scale parameter, usually taken as 30 s.

[0141] ETA — Estimated Time of Arrival, in seconds (s).

[0142] exp(⋅) — the natural exponential function.

[0143] This correction function amplifies the threat of targets with shorter proximity times, consistent with the staff's intuitive judgment.

[0144] The purpose of step S33 is to consider the dynamic evolution of the target—the threat of drones continuously approaching the protected target should increase over time, giving the threat assessment time-domain sensitivity.

[0145] The purpose of step S3 is to conduct a quantitative threat assessment of each tracked UAV target based on a unified air situation map, calculate the threat index, and provide a basis for target priority ranking and interception resource allocation in step S4 task allocation.

[0146] Step S4: Based on the threat assessment results, estimate the success rate of each interception device against different targets, quantify the interception cost of each interception device, and perceive the current occupancy status and resource availability of each device in real time to determine the availability of the devices.

[0147] In this embodiment, step S4 may specifically include the following steps: S41. Based on the maximum effective interception distance, minimum effective interception distance, interception azimuth range, maximum speed of interceptable targets, and target type adaptability of each interception equipment, the success probability of interception of each equipment against different targets is calculated comprehensively. When the target exceeds the envelope, the success probability of interception is zero.

[0148] Successful interception probability P int (p,j) consists of distance factor, angle factor, velocity factor, and target type adaptation factor: P int (p,j)=P range ⋅P angle ⋅P speed ⋅P match .

[0149] Where: P int (p,j) — The probability of successfully intercepting target j by equipment p (dimensionless, value range [0,1]).

[0150] P range — Distance factor (dimensionless, value range [0,1]).

[0151] P angle — Angle factor (dimensionless, with a value of 0 or 1).

[0152] P speed — Velocity factor (dimensionless, value range [0,1]).

[0153] P match — Target type adaptation factor (dimensionless, value range [0,1]).

[0154] The factors are defined as follows: Distance factor P range Based on the current distance r of the target j Maximum effective interception distance of the equipment Minimum effective interception distance Relationship: .

[0155] Where: r j — The slant distance of target j from equipment p, in meters (m).

[0156] — The minimum effective interception range of equipment p, in meters (m).

[0157] — The maximum effective interception range of equipment p, in meters (m).

[0158] — The optimal interception range for equipment p. The unit is meters (m).

[0159] |⋅| — Absolute value operation.

[0160] Angle factor P range Considering the firing / interception azimuth range of the equipment [θ] min ,θ max When the target azimuth angle θ j If the value is within this range, take 1; otherwise, take 0.

[0161] velocity factor P speed Based on target velocity v j With the equipment capable of intercepting the target's maximum speed The ratio: .

[0162] Where: v j — The current velocity of target j, in meters per second (m / s).

[0163] — Equipment p represents the maximum speed at which a target can be intercepted, measured in meters per second (m / s).

[0164] Target type fit factor P match The hard-kill weapon's effectiveness against large targets (P) is determined by referring to a table based on the matching degree between the equipment's interception capability and the target type (e.g., multi-rotor, fixed-wing, swarm). match High, soft-kill jamming equipment against swarm targets P match high.

[0165] The purpose of step S41 is to quantify the success rate of each interceptor equipment in intercepting the target, which depends on the matching degree between the equipment envelope (range, altitude, guidance method, and kill mechanism) and the target status.

[0166] S42 breaks down the interception cost into three components: ammunition or energy consumption cost, equipment occupation time cost, and collateral damage risk cost. After normalizing each component, the weighted sum is obtained to obtain the comprehensive interception cost. The comprehensive cost of soft-kill jamming is lower than that of hard-kill jamming.

[0167] The interception cost C(p) includes the cost of ammunition / energy consumption C. ammo Equipment occupancy time cost C occupy Collateral damage risk cost C collat : C(p) = β1⋅C ammo +β2⋅C occupy +β3⋅C collat .

[0168] Where: C(p) — the total cost of equipment p performing one interception (dimensionless, normalized).

[0169] C ammo — Cost of ammunition or energy consumption (in raw units, such as dollars or kilowatt-hours).

[0170] C occupy — Equipment occupancy time cost (raw units, such as US dollars).

[0171] C collat — Cost of incidental damage risk (in original units, such as dollars or risk scores).

[0172] β1, β2, β3 — the weighting coefficients corresponding to each item, satisfying β1+β2+β3=1.

[0173] Ammunition / Energy Consumption Cost C ammo Hard-kill missiles have a unit price (e.g., C). missile =1 million US dollars), the cost of laser weapons is the converted cost of launch energy consumption, and the cost of radio frequency interference is the cost of electrical energy consumption. According to the principle of prioritizing low-cost interference, the C of soft-kill means ammo It is typically 1-2 orders of magnitude lower than hard kill.

[0174] Equipment occupancy time cost C occupy The opportunity cost that the equipment cannot perform other tasks during the interception mission, in addition to the duration Δt of the interception process. engage Proportional: C occupy =η p ⋅Δt engage Wherein: η p — The unit time cost coefficient of equipment p. Δ tengage — Duration of the interception process, in seconds (s).

[0175] Collateral damage risk cost C collat The risk of collateral damage to surrounding airspace and ground facilities that may result from using this interception method is higher for hard-kill methods (such as missile fragments) and lower for soft-kill methods (such as navigation deception).

[0176] The three components have different dimensions, so they need to be normalized before weighted summation. Min-max normalization is used for normalization. .

[0177] in: — The normalized value of the kth cost item (dimensionless, range [0,1]).

[0178] C k — The original value of the kth cost item.

[0179] — The minimum value of the k-th cost component among all equipment.

[0180] — The maximum value of the k-th cost component among all equipment.

[0181] The purpose of step S42 is to quantify the interception cost of each interception device, distinguish between low-cost jamming methods and high-cost hard-kill methods, and provide a basis for cost-performance trade-off optimization.

[0182] S43 receives real-time information on the occupancy status, remaining ammunition or energy ratio, system health status, and cooling status reported by each interception equipment via a data link. Only when the equipment is idle, its resource reserves are not lower than the minimum threshold required to perform an interception, and its system health status is normal, is the equipment determined to be available and included in the task allocation candidate set.

[0183] The data link established in step S1 is used to receive status messages reported by each piece of equipment in real time, including: Occupied State S busy ∈{0,1}: 0 indicates idle, 1 indicates that an interception task is being executed (including reload time); Remaining Ammunition / Energy R resource ∈[0,1]: Normalized proportion of remaining resources; System health status S health ∈{0,1}: 0 indicates failure / unavailable, 1 indicates normal; Cooling conditions: Laser weapons need to take into account the cooling time between firing intervals, and radio frequency interference needs to take into account the heat dissipation recovery time.

[0184] The condition for determining equipment availability is: Available(p) = (S busy =0) ∧ (R resource ≥ ) ∧(S health =1).

[0185] Where: Available(p) — Available flag for equipment p (logical value, true or false).

[0186] S busy — The occupancy status of equipment p, 0 for idle, 1 for currently executing a task.

[0187] R resource — The remaining resource percentage of equipment p (dimensionless, [0,1]).

[0188] — The minimum resource threshold (dimensionless) required for equipment p to perform one interception.

[0189] S health — The system health status of equipment p, 0 indicates a fault, and 1 indicates normal operation.

[0190] ∧ — Logical AND operator.

[0191] Only equipment with Available(p)=True is included in the task assignment candidate set.

[0192] The purpose of step S43 is to obtain the current occupancy status and ammunition / energy reserves of each interception equipment, determine whether they can be used for new mission allocation, and ensure the feasibility of the allocation result.

[0193] The purpose of step S4 is to model the success probability and cost of interception for each interception equipment against different targets, provide input parameters for task allocation in step five, and support an allocation strategy that prioritizes low-cost jamming and reserves high-cost hard kill for backup.

[0194] Step S5: Based on the target threat index and the effectiveness, cost and availability parameters of the interception equipment, establish a multi-constraint multi-target task allocation optimization model, solve the optimal assignment scheme of detection and interception tasks, and generate the execution instruction sequence of each piece of equipment.

[0195] In this embodiment, step S5 may specifically include the following steps: S51 constructs a task allocation mathematical model with the dual optimization objectives of maximizing overall interception effectiveness and minimizing total interception cost. The threat index of each target is used as the effectiveness weight, and the probability of successful interception and the comprehensive interception cost are used as decision parameters. The constraints are set as follows: each target must be allocated at least one piece of equipment, each piece of equipment can intercept at most one target at the same time, equipment capability matching, and the upper limit of the number of pieces of equipment allocated to the same target.

[0196] Suppose there are M interception equipment (firepower units) and N targets (drones), define the allocation decision variable x. ij x ∈{0,1} ij =1 indicates that equipment i is assigned to target j, x ij =0 indicates no allocation.

[0197] Construct a dual-objective optimization model: Objective 1 (Maximize overall interception effectiveness): .

[0198] Where: F1 — Overall interception effectiveness (dimensionless, the larger the better).

[0199] N — Total number of targets in the current region.

[0200] M — Total number of available interception equipment.

[0201] P ij — The probability (dimensionless) of successful interception of target j by equipment i is calculated by step S41.

[0202] x ij — Binary decision variables, 1 indicates that equipment i will be assigned to target j, and 0 indicates that it will not be assigned.

[0203] w j — The normalized threat weight of target j is transformed from the threat index TI in step S3, w j =TI j / ∑ k=1 N TI k .

[0204] TI j — Threat index of target j (calculated by step S3).

[0205] ∏ — Multiplication operation.

[0206] ∑ — Summation operation.

[0207] The physical meaning of this objective function is: for each target, the combined success probability of interception by all assigned equipment is 1 - ∏(1 - P) / ... ij The probability of a successful interception by at least one piece of equipment is calculated by multiplying the probability by the target's threat weight and summing the results. This encourages prioritizing resources for high-threat targets and allows multiple pieces of equipment to be assigned to the same target to improve the interception success rate.

[0208] Objective 2 (Minimize total interception cost): .

[0209] Where: F2 — Total interception cost (dimensionless, the smaller the better).

[0210] C ij — The total cost of equipment i intercepting target j (calculated by step S4).

[0211] There is a conflict between the two objectives—pursuing a high success rate may require allocating multiple pieces of equipment or using costly hard-kill methods, leading to increased costs; pursuing low costs may involve allocating only a single soft-kill method, resulting in a lower success rate. Therefore, this is a multi-objective optimization problem, requiring the solution to the Pareto optimal solution set.

[0212] The constraints include: (1) Allocate at least one piece of equipment to each target (if resources are sufficient) or as much as possible (if the number of equipment is less than the number of targets): , ∀j=1,2,...,N (if M≥N).

[0213] Wherein: ∀ — for all (universal quantifiers).

[0214] (2) Each piece of equipment can intercept a maximum of one target at a time: , ∀i=1,2,...,M.

[0215] (3) Equipment capability constraint: x only applies when target j is within the envelope of equipment i. ij Only 1 is allowed for x ij ≤Capability ij Capability ij ∈{0,1} indicates whether equipment i can intercept target j, where 1 indicates yes and 0 indicates no, and is determined by the interception condition in step S41.

[0216] (4) Equipment quantity limit constraint: The number of equipment allocated to the same target shall not exceed the allowable limit L. m : , ∀j=1,2,...,N.

[0217] Among them, L m — The maximum number of equipment allocated to the same target is usually 2 to 3 to avoid excessive waste.

[0218] This constraint setting ensures that the allocation strategy of using low-cost jamming methods to deal with low-threat targets and high-cost hard-kill methods to deal with high-threat targets can be mathematically implemented—high-threat targets are assigned a larger w in the objective function. j Low-threat target w j Smaller targets naturally tend to be prioritized for low-cost equipment allocation, leading to a better optimal solution.

[0219] The purpose of step S51 is to formalize the task allocation problem into a mathematical model and clarify the optimization objective and constraints.

[0220] S52 employs a non-dominated sorting genetic algorithm with a constraint repair mechanism to solve the bi-objective optimization model. The allocation scheme is represented by binary encoding, and population diversity is maintained through non-dominated sorting and crowding distance calculation. Population evolution is carried out through selection, crossover, and mutation operations. Gene positions of individuals that do not meet the ability constraints are forcibly corrected, and finally, the Pareto optimal solution set is output.

[0221] An improved non-dominated sorting genetic algorithm with a repair mechanism (NSGA-II) is adopted.

[0222] (1) Encoding and Population Initialization: Decision variable matrix X=[x ij ] M×N Binary encoding is used, with each individual representing an allocation scheme. The population size is set to N. pop =100, initialize the population by randomly generating feasible solutions that satisfy the constraints. For gene bits that do not satisfy the equipment capability constraints, initialize to 0 and mark them as fixed bits.

[0223] (2) Calculation of non-dominated ordering and crowding distance: For each individual in the population, Pareto dominance is determined according to objective functions F1 and F2. Individual p dominates individual q if and only if: F1(p) ≥ F1(q) and F2(p) ≤ F2(q), and at least one inequality holds strictly. All individuals are stratified into different non-dominated levels (Rank 1 is the optimal Pareto front). Within the same level, individuals are ranked according to their crowding distance d. i Sort, d i Defined as the sum of the distances of individual i to its two nearest neighbors in the target space, used to maintain population diversity.

[0224] (3) Genetic manipulation and repair mechanisms: The selection operation uses tournament selection (selecting the better individual from two randomly chosen individuals to enter the mating pool). The crossover operation uses two-point crossover, and the mutation operation uses bit-flip mutation (mutation rate pm=0.01).

[0225] Crossover and mutation may result in Capability that does not meet the constraints. ij For individuals, a repair mechanism is introduced: for each gene locus x that violates the ability constraint ij =1 but Capability ij =0, force it to be corrected to 0; for targets that have not been assigned any equipment (∑ i x ij=0), randomly select an allocation from the equipment capable of intercepting the target to ensure that the constraint is satisfied.

[0226] (4) Elite retention and iteration termination: An elite retention strategy is adopted, and after merging the parent and child generations, the top N generations are selected based on non-dominance sorting and crowding. pop Each individual enters the next generation. The iteration termination condition is set to the maximum number of iterations G. max =200 or the Pareto frontal change rate is less than the threshold ϵ=10 for 50 consecutive generations. -4 .

[0227] The final output is a set of Pareto optimal solutions, each solution corresponding to a non-dominated trade-off between interception effectiveness and interception cost, which staff can choose based on the current situation.

[0228] The purpose of step S52 is to solve the Pareto optimal solution set of the above bi-objective optimization model, achieving a balance between computational efficiency and solution quality.

[0229] S53 selects the optimal allocation scheme from the Pareto optimal solution set, sorts multiple interception tasks of the same equipment according to the order of the expected arrival time of the target in the kill zone, verifies whether the time interval between consecutive tasks of the same equipment meets the minimum reload or cooldown time requirements, generates the interception command execution sequence of each equipment and issues it for execution.

[0230] For the optimal allocation scheme in the Pareto optimal solution set (selected based on the commander's decision preference), a time window scheduling model is constructed: Each piece of equipment i was assigned a target set J. i (If ∑ j x ij =0 indicates an empty value. For each allocation (i,j), calculate the expected interception time. (Starting from the current moment, consider the missile flight time or laser illumination time). The constraint is that a time interval must be satisfied between two consecutive interception missions by the same equipment. (Including loading / cooling time): , ∀j1≠j2, j1,j2∈J i .

[0231] in: — The estimated interception time of target j by equipment i, in seconds (s).

[0232] — Equipment i The minimum time interval (including reload / cooldown) required between two consecutive intercept missions, in seconds (s).

[0233] Ji — The set of targets assigned to equipment i.

[0234] |⋅| — Absolute value operation.

[0235] ∀ — For all.

[0236] This is a single-machine scheduling problem, which uses the earliest deadline first (EDF) heuristic algorithm for sorting: interception tasks are executed in the order of the targets' expected arrival time in the kill zone (calculated by step S3 ETA), with higher priority targets being scheduled first.

[0237] The final output is an interception command sequence for each piece of equipment, including target ID, interception time, and interception method (soft / hard kill parameters), which is sent to each piece of equipment for execution through the data link established in step S1.

[0238] The purpose of step S53 is to transform the static allocation scheme obtained by NSGA-II into actual scheduling instructions that take into account the execution timing of each piece of equipment, so as to ensure that multiple interception tasks do not conflict in time.

[0239] The purpose of step S5 is to dynamically assign detection and tracking tasks and interception and strike tasks to the most suitable nodes based on the target threat assessment results of step S3 and the equipment effectiveness-cost parameters of step S4, using a multi-constraint task allocation algorithm, thereby realizing the sharing of detection resources and intelligent allocation of interception firepower.

[0240] Step S6 involves real-time monitoring of situational changes during task execution. When triggering conditions are met, rapid replanning is initiated to reassign unfinished interception tasks. Subsequent strategies are then corrected based on the interception effect evaluation, forming a closed loop of detection-evaluation-decision-execution-feedback.

[0241] In this embodiment, step S6 may specifically include the following steps: S61 monitors in real time the changes in the status of each target, the appearance of new targets and the disappearance of existing targets, as well as the changes in the occupation and resource status of each interception equipment. When the change of any monitored quantity exceeds the preset threshold, the task replanning process is triggered.

[0242] The design incorporates three levels of trigger conditions; replanning is initiated when any one condition is met. (1) Significant changes in target state: The changes in the position, velocity, or heading of any target exceed the threshold, defined as: Δ target =(∥p j (k)-p j (k-Δt)∥) / Δt+λ v ∥v j (k)-v j (k-Δt)∥, if Δtarget >θ trigger If the rate of change of position exceeds 50 m / s or the velocity vector change exceeds 30°, a replanning is triggered.

[0243] Where: Δ target — Comprehensive index of target state change (unit: m / s, combining position change rate and velocity change).

[0244] p j (k) — The position vector of target j at time k (in meters).

[0245] p j (k-Δt) — The position vector of target j at the previous time k-Δt (in meters).

[0246] Δt — Time interval (in seconds).

[0247] v j (k) — The velocity vector of target j at time k (in m / s).

[0248] v j (k-Δt) — The velocity vector of target j at the previous time k-Δt (in m / s).

[0249] ||⋅|| — The Euclidean norm (length) of a vector. λ v — The weighting coefficient (dimensionless) of the velocity change term, used to balance the contributions of position change and velocity change.

[0250] (2) New target appears or target disappears: Step S4 track association discovers a new track ID or an existing track is determined to have disappeared if it has not been updated for 3 consecutive detection cycles, triggering replanning.

[0251] (3) Equipment status change: Step S4 detects that an interception device is occupied, damaged, has exhausted its ammunition, or has been released after completing its current mission, triggering a replanning. In particular, if an interception fails (the target is confirmed not to have been destroyed through kill assessment), the target needs to be reassigned, triggering an emergency replanning.

[0252] The purpose of step S61 is to monitor changes in the situation in real time, determine whether task replanning is triggered, and avoid frequent replanning that could lead to wasted computing resources and instruction confusion.

[0253] S62 takes the actual target state and equipment state at the trigger time of replanning as the initial conditions, optimizes only the task allocation scheme within the preset time window to shorten the replanning solution time, and adopts a hot-start strategy that uses the optimization results of the previous round as part of the initial population to accelerate the convergence of the algorithm.

[0254] A receding horizontal optimization strategy is adopted. The replanning trigger time is set to t0, and optimization is only performed on the future time T. window Task allocation within a time window (usually T) window =30 s), tasks outside the window will either continue with the original plan or remain undecided. The calculation process from steps S1 to S5 is compressed into a fast replanning module, using the current target state and equipment state as initial inputs, and running a simplified NSGA-II (population size reduced to N) pop =30, the maximum number of iterations is reduced to G. max =50) at t replan Output a new T within ≤2 seconds window Window allocation scheme.

[0255] To improve the efficiency of online solution, a warm start strategy can be introduced: the Pareto optimal solution from the previous round of optimization is used as part of the initial population (accounting for 20%), which accelerates the convergence of the algorithm.

[0256] The purpose of step S62 is to quickly solve the redistribution scheme of the remaining tasks after triggering replanning, using the current actual state as the initial condition, avoiding global optimization from scratch and improving response speed.

[0257] S63, within a preset evaluation time after each interception mission, confirms the target area through the detection system. If the target track disappears, the interception is deemed successful and the corresponding equipment resources are released. If the target track still exists, the interception is deemed unsuccessful, the target is re-added to the allocation queue and its threat index is increased, triggering the replanning process to reallocate interception resources. At the same time, the interception effect statistics are fed back to the interception success probability model for parameter correction.

[0258] After each interception task is executed, t assessment Within a certain timeframe (usually 5 seconds after the target is expected to be hit), the detection system performs a confirmation detection of the target area: If the target track disappears and no new track appears, the interception is considered successful, and the equipment resources that performed the mission are released. If the target's trajectory still exists and its motion state does not change significantly, the interception is deemed to have failed. The target is marked as not destroyed, re-included in the target queue, and its threat index is increased (multiplied by a coefficient of 1.5), triggering the replanning process in step S62 to reallocate interception resources. If the target track breaks apart or fragmentation signals appear, further analysis is needed to determine whether it is a decoy release or target disintegration, with staff assisting in the decision-making process.

[0259] The results of the interception effectiveness evaluation are fed back to the trajectory fusion module in step S2 and the task allocation model in step S5, forming a complete closed-loop learning mechanism—long-term statistical analysis of the actual interception success rate of each piece of equipment against different types of targets, used to dynamically correct the P in step S4. int The parameters allow the performance model to gradually approximate the real battlefield environment.

[0260] The purpose of step S63 is to: confirm the execution effect of each interception task through kill assessment, re-include unintercepted targets in the allocation queue, and correct subsequent interception strategies.

[0261] The purpose of step S6 is to adjust the task allocation plan online when the situation changes dynamically (target maneuvering, new targets appearing, friendly equipment being damaged or occupied, interception effect not meeting expectations), forming a closed-loop feedback of detection-assessment-decision-execution-re-detection to ensure the continuous effectiveness of the system.

[0262] This invention can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This invention can be described in the general context of computer-executable instructions, such as program modules, that are executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This invention can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0263] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by instructing related hardware through computer-readable instructions. These computer-readable instructions can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above methods. The aforementioned storage medium can be a non-volatile storage medium such as a magnetic disk, optical disk, or read-only memory (ROM), or random access memory (RAM).

[0264] It should be understood that although the steps in the flowcharts of the accompanying figures are shown sequentially as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the accompanying figures may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times, and their execution order is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the sub-steps or stages of other steps.

[0265] Example 2 Further reference Figure 2 As a response to the above Figure 1 The present invention provides an embodiment of a cooperative detection and task allocation device in an anti-drone air defense system, which is implemented by the method shown. Figure 1 Corresponding to the method embodiments shown, this device can be specifically applied to various electronic devices.

[0266] like Figure 2 As shown, the collaborative detection and task allocation device 70 in the anti-drone air defense system described in this embodiment includes: a setup module 71, an identification module 72, an extraction module 73, a quantization module 74, a setup module 75, and a correction module 76. Wherein: Module 71 is established to unify the access of dispersed radar detection nodes, photoelectric detection nodes, radio frequency detection nodes and various soft and hard kill interception equipment in the region to the collaborative command and control platform, establish a unified time reference and spatial coordinate system, and realize the interconnection and interoperability of heterogeneous equipment and the standardized aggregation of data. The identification module 72 is used to perform point-track association on the aggregated multi-source heterogeneous detection data, identify heterogeneous sensor measurements belonging to the same target, re-associate interrupted track segments in the radar blind zone, and generate a unified air situation map through data fusion. The extraction module 73 is used to extract the motion characteristics, attack intentions and physical attributes of each tracked target based on the generated unified air situation map, quantify multi-dimensional threat factors, comprehensively calculate the threat index of each target, and perform threat time domain correction for targets that are continuously approaching the protected target. The quantification module 74 is used to estimate the success probability of each interception device against different targets based on the threat assessment results, quantify the interception cost of each interception device, and perceive the current occupancy status and resource availability of each device in real time to determine the availability of the devices. Module 75 is established to build a multi-constraint, multi-target task allocation optimization model based on the target threat index and the effectiveness, cost, and availability parameters of interception equipment, solve the optimal assignment scheme for detection and interception tasks, and generate the execution instruction sequence for each piece of equipment. The correction module 76 is used to monitor situational changes in real time during task execution. When the triggering conditions are met, it initiates rapid replanning, reallocates unfinished interception tasks, and corrects subsequent strategies through interception effect evaluation, forming a closed loop of detection-evaluation-decision-execution-feedback.

[0267] Example 3 To address the aforementioned technical problems, embodiments of the present invention also provide a computer device. Please refer to [link / reference needed]. Figure 3 , Figure 3 This is a basic structural block diagram of the computer device in this embodiment.

[0268] The aforementioned computer device 8 includes a memory 81, a processor 82, and a network interface 83 that are interconnected via a system bus. It should be noted that only the computer device 8 with components 81, 82, and 83 is shown in the figure; however, it should be understood that it is not required to implement all the shown components, and more or fewer components can be implemented alternatively. Those skilled in the art will understand that the computer device described herein is a device capable of automatically performing numerical calculations and / or information processing according to pre-set or stored instructions, and its hardware includes, but is not limited to, microprocessors, application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), digital signal processors (DSPs), embedded devices, etc.

[0269] The aforementioned computer devices can be desktop computers, laptops, handheld computers, and cloud servers, among other computing devices. These devices can facilitate human-computer interaction with users through keyboards, mice, remote controls, touchpads, or voice-activated devices.

[0270] The aforementioned memory 81 includes at least one type of readable storage medium, including flash memory, hard disk, multimedia card, card-type memory (e.g., SD or DX memory), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, magnetic disk, optical disk, etc. In some embodiments, the aforementioned memory 81 may be an internal storage unit of the aforementioned computer device 8, such as the hard disk or memory of the computer device 8. In other embodiments, the aforementioned memory 81 may also be an external storage device of the aforementioned computer device 8, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the computer device 8. Of course, the aforementioned memory 81 may also include both the internal storage unit and its external storage device of the aforementioned computer device 8. In this embodiment, the aforementioned memory 81 is typically used to store the operating system and various application software installed on the aforementioned computer device 8, such as computer-readable instructions for collaborative detection and task allocation methods under an anti-drone air defense system. In addition, the aforementioned memory 81 can also be used to temporarily store various types of data that have been output or will be output.

[0271] In some embodiments, the processor 82 may be a central processing unit (CPU), controller, microcontroller, microprocessor, or other data processing chip. The processor 82 is typically used to control the overall operation of the computer device 8. In this embodiment, the processor 82 is used to execute computer-readable instructions stored in the memory 81 or to process data, such as executing computer-readable instructions for the cooperative detection and task allocation method under the anti-drone air defense system.

[0272] The network interface 83 may include a wireless network interface or a wired network interface, which is typically used to establish a communication connection between the computer device 8 and other electronic devices.

[0273] Example 4 The present invention also provides another embodiment, namely, a computer-readable storage medium storing computer-readable instructions that can be executed by at least one processor to cause the at least one processor to perform the steps of the cooperative detection and task allocation method under the above-described anti-drone air defense system.

[0274] The beneficial effects of implementing the above embodiments are as follows: (1) Adopting a systematic collaborative architecture to upgrade single-equipment independent combat to network-based intelligent combat: Through networking and the establishment of a unified benchmark, the dispersed radar, photoelectric, radio frequency detection nodes and various interception equipment are transformed from information islands into resource pools that can be collaboratively scheduled, realizing the sharing of detection resources and the relay of interception tasks.

[0275] (2) Multi-source track association and fusion: Through hierarchical clustering and Hausdorf distance reassociation technology, the problem of difficulty in re-acquiring targets lost in the radar blind zone (below 150 meters) has been solved, and continuous tracking and maintenance under multi-node relay detection has been achieved, which significantly expands the effective coverage of the defense system.

[0276] (3) Multi-constraint task allocation: Through threat assessment, effectiveness cost modeling, and multi-objective optimization, low-cost interference means are prioritized to deal with low-threat targets, while high-cost hard-kill means are reserved to deal with high-threat targets, thus optimizing the overall defense cost-effectiveness ratio.

[0277] (4) Dynamic replanning closed-loop optimization: Through real-time situation detection and rolling optimization, the system can adapt to dynamic changes in the situation (swarm attack, simultaneous attack from multiple directions, target maneuvering and evasion), which improves the success rate of defense against sudden scenarios.

[0278] Through the above description of the embodiments, those skilled in the art can clearly understand that the methods of the above embodiments can be implemented by means of software plus necessary general-purpose hardware platforms. Of course, they can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) and includes several instructions to cause a terminal device (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods of the various embodiments of the present invention.

[0279] Obviously, the embodiments described above are merely some embodiments of the present invention, not all embodiments. The accompanying drawings show preferred embodiments of the present invention, but do not limit the patent scope of the present invention. The present invention can be implemented in many different forms; rather, these embodiments are provided to provide a more thorough and complete understanding of the disclosure of the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing specific embodiments, or make equivalent substitutions for some of the technical features. Any equivalent structures made using the content of this specification and drawings, directly or indirectly applied to other related technical fields, are similarly within the patent protection scope of this invention.

Claims

1. A method for collaborative detection and task allocation in an anti-drone air defense system, characterized in that, Includes the following steps: The platform integrates the dispersed radar detection nodes, photoelectric detection nodes, radio frequency detection nodes, and various soft and hard kill interception equipment within the region into a unified collaborative command and control platform, establishing a unified time reference and spatial coordinate system to achieve interconnection and interoperability of heterogeneous equipment and standardized data aggregation. The system performs point-track-track association on the aggregated multi-source heterogeneous detection data, identifies heterogeneous sensor measurements belonging to the same target, re-associates interrupted track segments within radar blind zones, and generates a unified air situation map through data fusion. Based on the generated unified air situation map, the motion characteristics, attack intentions and physical attributes of each tracked target are extracted, multi-dimensional threat factors are quantified, the threat index of each target is calculated in a comprehensive manner, and the threat time domain correction is performed on targets that continue to approach the protected target. Based on the threat assessment results, the success probability of each interception device against different targets is estimated, the interception cost of each interception device is quantified, and the current occupancy status and resource availability of each device are perceived in real time to determine the availability of the devices. Based on the target threat index and the effectiveness, cost and availability parameters of interception equipment, a multi-constraint multi-target task allocation optimization model is established to solve the optimal assignment scheme of detection and interception tasks and generate the execution instruction sequence of each piece of equipment. During mission execution, the system monitors situational changes in real time. When triggering conditions are met, it initiates rapid replanning, reassigns unfinished interception tasks, and corrects subsequent strategies through interception effectiveness evaluation, forming a closed loop of detection-evaluation-decision-execution-feedback.

2. The method for collaborative detection and task allocation in an anti-drone air defense system according to claim 1, characterized in that, The steps of unifying the dispersed radar detection nodes, photoelectric detection nodes, radio frequency detection nodes, and various soft and hard kill interception equipment within the region into the collaborative command and control platform, establishing a unified time reference and spatial coordinate system, and realizing the interconnection and interoperability of heterogeneous equipment and the standardized aggregation of data specifically include: Through a layered access adaptation architecture, corresponding physical access modules are configured for the communication interfaces of different equipment. The private communication protocols of each piece of equipment are uniformly converted into standard internal data exchange protocols through protocol conversion middleware. Each equipment node registers to the collaborative command and control platform through a standardized agent, and the platform maintains the global equipment registry. The BeiDou satellite timing module is used as a unified time source. BeiDou timing receivers are deployed at each access node. Time synchronization calibration is performed in the local area network through a precise time protocol. The recursive least squares algorithm is used to dynamically compensate for clock drift, so that all nodes in the network maintain a unified time reference. Each data packet after synchronization is accompanied by a globally unified timestamp. A geocentric rectangular coordinate system or a station-centered coordinate system is selected as a common reference coordinate system. The measurement data of each detection node under different local coordinate systems are transformed to the common reference coordinate system, and the measurement error after coordinate transformation is estimated by the error propagation analysis method.

3. The method for collaborative detection and task allocation in an anti-drone air defense system according to claim 1, characterized in that, The steps of performing point-track-track association on the aggregated multi-source heterogeneous detection data, identifying heterogeneous sensor measurements belonging to the same target, re-associating interrupted track segments within radar blind zones, and generating a unified air situation map through data fusion specifically include: For the established target trajectory, the predicted state and predicted covariance of the trajectory are obtained through filtering algorithm. The normalized innovation distance between the newly arrived detection point trajectory and each trajectory is calculated. The association decision is made by elliptical association gate. A stop-and-go model is added to the motion model to adapt to the hovering and low-speed maneuvering characteristics of low, slow and small targets. Motion feature vectors of each track segment before and after the interruption are extracted. The Hausdorff distance between track segments is used as a similarity measure. A bottom-up agglomerative hierarchical clustering method is used to re-associate interrupted track segments belonging to the same target. During the clustering process, the consistency of motion features is constrained. Multi-sensor measurements associated with the same target are fused, and the inverse matrix of the covariance matrix of each sensor's state estimate is used as the information weight. The state estimates of each sensor are weighted and summed to obtain the fused target state and its covariance, generating a unified air situation map.

4. The method for collaborative detection and task allocation in an anti-drone air defense system according to claim 1, characterized in that, The steps of extracting the motion characteristics, attack intent, and physical attribute features of each tracked target based on the generated unified air situation map, quantifying multi-dimensional threat factors, comprehensively calculating the threat index of each target, and performing threat time-domain correction for targets continuously approaching the protected target specifically include: Normalized threat factors are extracted from five dimensions: target speed, target heading towards the degree of protection of the target, target flight altitude, target radar cross-section, and number of targets. The five threat factors are used to form a factor vector. The weight vector of each factor is determined by the analytic hierarchy process. Each factor is mapped to a preset evaluation set by the membership function. A fuzzy relation matrix is ​​constructed and a fuzzy comprehensive synthesis operation is performed to obtain the comprehensive threat index of each target. The estimated arrival time of each target to the protected target is calculated, and a time-domain correction function is constructed based on the estimated arrival time to dynamically correct the comprehensive threat index, so that the threat index of targets that continue to approach the protected target increases over time.

5. The method for collaborative detection and task allocation in an anti-drone air defense system according to claim 1, characterized in that, The steps of estimating the success probability of interception for each interceptor against different targets based on the threat assessment results, quantifying the interception cost of each interceptor, and real-time sensing of the current occupancy status and resource availability of each device to determine equipment availability specifically include: Based on the maximum effective interception distance, minimum effective interception distance, interception azimuth range, maximum speed of interceptable targets, and target type adaptability of each interception equipment, the success probability of interception of each equipment against different targets is calculated comprehensively. When the target exceeds the envelope, the success probability of interception is zero. Interception costs are broken down into three components: ammunition or energy consumption costs, equipment occupation time costs, and collateral damage risk costs. Each component is normalized and then weighted and summed to obtain the comprehensive interception cost. The comprehensive cost of soft-kill jamming methods is lower than that of hard-kill methods. The system receives real-time reports from each interception device regarding its occupancy status, remaining ammunition or energy percentage, system health status, and cooling status via a data link. The device is deemed available and included in the task allocation candidate set only when it is idle, has sufficient resources (not lower than the minimum threshold required to perform an interception), and its system health status is normal.

6. The method for collaborative detection and task allocation in an anti-drone air defense system according to claim 1, characterized in that, The steps of establishing a multi-constraint, multi-target task allocation optimization model based on the target threat index and the effectiveness, cost, and availability parameters of interception equipment, solving for the optimal assignment scheme of detection and interception tasks, and generating the execution command sequence of each piece of equipment specifically include: A mathematical model for task allocation is constructed with the dual optimization objectives of maximizing overall interception effectiveness and minimizing total interception cost. The threat index of each target is used as the effectiveness weight, and the probability of successful interception and the overall interception cost are used as decision parameters. The constraints are set as follows: each target must be allocated at least one piece of equipment, each piece of equipment can intercept at most one target at the same time, equipment capability matching, and the upper limit of the number of pieces of equipment allocated to the same target. The bi-objective optimization model is solved by a non-dominated sorting genetic algorithm with a constraint repair mechanism. The allocation scheme is represented by binary encoding. Population diversity is maintained by non-dominated sorting and crowding distance calculation. Population evolution is carried out through selection, crossover and mutation operations. Gene positions of individuals that do not meet the ability constraints are forcibly corrected. Finally, the Pareto optimal solution set is output. The optimal allocation scheme is selected from the Pareto optimal solution set. For multiple interception tasks of the same equipment, the targets are sorted in order of their expected arrival time in the kill zone. The time interval between consecutive tasks of the same equipment is checked to see if it meets the minimum reload or cooldown time requirement. An interception command execution sequence for each equipment is generated and issued for execution.

7. The method for collaborative detection and task allocation in an anti-drone air defense system according to any one of claims 1 to 6, characterized in that, The steps of real-time monitoring of situational changes during task execution, initiating rapid replanning when triggering conditions are met, reallocating unfinished interception tasks, and correcting subsequent strategies through interception effect evaluation to form a closed loop of detection-evaluation-decision-execution-feedback specifically include: Real-time monitoring of changes in the status of each target, the appearance of new targets and the disappearance of existing targets, as well as changes in the occupation and resource status of each interception equipment. When the change of any monitored quantity exceeds the preset threshold, the task replanning process is triggered. Using the actual target state and equipment state at the trigger time of replanning as initial conditions, the task allocation scheme within the preset time window is optimized only to shorten the replanning solution time. A hot-start strategy that uses the optimization results of the previous round as part of the initial population is adopted to accelerate the convergence of the algorithm. Within a preset evaluation period after each interception mission, the target area is confirmed by the detection system. If the target track disappears, the interception is considered successful and the corresponding equipment resources are released. If the target track still exists, the interception is considered unsuccessful, the target is re-added to the allocation queue and its threat index is increased, triggering a replanning process to reallocate interception resources. At the same time, the interception effect statistics are fed back to the interception success probability model for parameter correction.

8. A collaborative detection and task allocation device for an anti-drone air defense system, characterized in that, include: A module is established to unify the access of dispersed radar detection nodes, photoelectric detection nodes, radio frequency detection nodes, and various soft and hard kill interception equipment to the collaborative command and control platform, establish a unified time reference and spatial coordinate system, and realize the interconnection and interoperability of heterogeneous equipment and the standardized aggregation of data. The identification module is used to perform point-track association on the aggregated multi-source heterogeneous detection data, identify heterogeneous sensor measurements belonging to the same target, re-associate interrupted track segments in the radar blind zone, and generate a unified air situation map through data fusion. The extraction module is used to extract the motion characteristics, attack intentions and physical attributes of each tracked target based on the generated unified air situation map, quantify multi-dimensional threat factors, comprehensively calculate the threat index of each target, and perform threat time-domain correction for targets that are continuously approaching the protected target. The quantification module is used to estimate the success probability of each interception device against different targets based on the threat assessment results, quantify the interception cost of each interception device, and perceive the current occupancy status and resource availability of each device in real time to determine the availability of the devices. A module is established to create a multi-constraint, multi-target task allocation optimization model based on the target threat index and the effectiveness, cost, and availability parameters of interception equipment. This model solves for the optimal assignment scheme of detection and interception tasks and generates the execution command sequence for each piece of equipment. The correction module is used to monitor situational changes in real time during task execution. When the triggering conditions are met, it initiates rapid replanning, reassigns unfinished interception tasks, and corrects subsequent strategies based on the interception effect evaluation, forming a closed loop of detection-evaluation-decision-execution-feedback.

9. A computer device, characterized in that, It includes a memory and a processor, wherein the memory stores computer-readable instructions, and the processor executes the computer-readable instructions to implement the steps of the cooperative detection and task allocation method under the anti-drone air defense system as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-readable instructions, which, when executed by a processor, implement the steps of the cooperative detection and task allocation method under the anti-drone air defense system as described in any one of claims 1 to 7.

Images