A computer-aided laser source control method and system

Abstract

This invention discloses a computer-aided laser source control method and system, relating to the field of laser source control technology. It addresses the problems of insufficient prediction accuracy and low interception efficiency in laser interception aiming. The method includes: acquiring current motion state data of a UAV and determining a prediction probability cloud based on this data; determining at least one error-correcting trajectory segment based on the UAV's historical trajectory data; determining the target accuracy of each predicted position in the prediction probability cloud based on the motion direction matching degree between the predicted position and each error-correcting trajectory segment; determining the damage expectation of each predicted position in the prediction probability cloud based on the target accuracy, the corresponding probability cloud density, and the UAV's attitude information at the predicted position; and determining an aiming point from the prediction probability cloud based on the damage expectations of all predicted positions, and controlling the laser source based on the aiming point.

Description

Technical Field

[0001] This invention relates to the field of laser source control technology, and specifically to a computer-aided laser source control method and system. Background Technology

[0002] With the rapid development of drone technology and its widespread application in military, security and other fields, the use of high-energy laser weapons for precise interception has become an important countermeasure. This technology relies on continuously focusing a high-energy laser beam on a specific part of a high-speed maneuvering drone target to achieve damage through thermal accumulation. The core challenge lies in how to achieve stable tracking of the beam on the target and efficient energy transfer.

[0003] In existing technologies, probabilistic cloud prediction methods are commonly used to address the uncertainty of target maneuverability. These methods predict the probability distribution of a target's future movement based on real-time sensor data (such as position and velocity), using a motion model to determine the laser's aiming point and power, aiming to balance energy delivery and tracking uncertainties. However, in complex real-world combat environments, the movement of UAVs is not only influenced by their current maneuverability but also closely related to their tactical intentions, historical behavior patterns, and environmental interactions. Existing methods primarily rely on the current instantaneous state for prediction, failing to adequately mine and utilize deeper information about the target's intentions. This results in a significant deviation between the generated predicted probability cloud and the target's actual movement, thus affecting the overall interception accuracy and combat effectiveness of laser weapon systems. Summary of the Invention

[0004] To address the technical problem of insufficient prediction accuracy and low interception efficiency in laser interception and targeting due to the current reliance on probabilistic cloud prediction based primarily on the instantaneous motion state of UAVs, which fails to effectively integrate the intent information contained in their historical trajectories, the present invention aims to provide a computer-aided laser source control method and system. The specific technical solution adopted is as follows: In a first aspect, the present invention provides a computer-aided laser source control method, comprising: acquiring current motion state data of a UAV and determining a prediction probability cloud based on the current motion state data; wherein the prediction probability cloud contains multiple prediction positions, each prediction position being associated with a probability cloud density predicted based on the current motion state of the UAV; determining at least one error correction trajectory segment based on historical trajectory data of the UAV; wherein the error correction trajectory segment is used to characterize the trajectory segment of the UAV after path replanning in historical flight to face an attack target; for each prediction position in the prediction probability cloud, determining the target accuracy of the prediction position based on the motion direction matching degree between the prediction position and each error correction trajectory segment; wherein the target accuracy is used to characterize the consistency between the prediction position and the UAV's attack target; for each prediction position in the prediction probability cloud, determining the damage expectation of the prediction position based on the target accuracy, the corresponding probability cloud density, and the attitude information of the UAV at the prediction position; determining an aiming point from the prediction probability cloud based on the damage expectations of all prediction positions, and controlling the laser source based on the aiming point.

[0005] Secondly, the present invention provides a computer-aided laser source control system, comprising: a prediction probability cloud module, a trajectory analysis module, an accuracy evaluation module, an expectation calculation module, and a source control module; the prediction probability cloud module is used to acquire the current motion state data of the UAV and determine the prediction probability cloud based on the current motion state data; wherein, the prediction probability cloud contains multiple prediction positions, and each prediction position is associated with a probability cloud density predicted based on the current motion state of the UAV; the trajectory analysis module is used to determine at least one error correction trajectory segment based on the historical trajectory data of the UAV; wherein, the error correction trajectory segment is used to characterize the UAV's trajectory towards the attack target during historical flight. The system comprises: a trajectory segment after path replanning; an accuracy assessment module, used to determine the target accuracy of each predicted position in the prediction probability cloud based on the motion direction matching degree between the predicted position and each error-corrected trajectory segment; where target accuracy characterizes the consistency between the predicted position and the UAV's attack target; an expectation calculation module, used to determine the damage expectation of each predicted position in the prediction probability cloud based on the target accuracy, the corresponding probability cloud density, and the UAV's attitude information at the predicted position; and a light source control module, used to determine the aiming point from the prediction probability cloud based on the damage expectations of all predicted positions, and control the laser light source based on the aiming point.

[0006] Thirdly, the present invention provides an electronic device, comprising: a processor and a memory; wherein the memory is used to store one or more programs, the one or more programs including computer-executable instructions, and when the electronic device is running, the processor executes the computer-executable instructions stored in the memory to cause the electronic device to perform the computer-aided laser source control method as described in the first aspect and any possible implementation thereof.

[0007] This invention offers the following advantages: by integrating real-time UAV motion prediction, historical intent trajectory analysis, and damage physics modeling, it achieves intelligent selection of laser aiming points and precise power control. Its technical effects are as follows: First, by analyzing "error correction" segments in historical trajectories, the true attack intent direction of the UAV is extracted, thereby correcting the reliability of probability clouds based solely on instantaneous state predictions, significantly improving the consistency between predicted positions and actual target directions. Second, by introducing an attitude damage factor, the physical attitude of the UAV when irradiated is incorporated into the evaluation system, ensuring that the final selected aiming point is not only a high-probability hit point but also a high-energy-efficiency damage point. Finally, based on comprehensive damage expectations, the aiming point is determined and the required power is dynamically calculated, achieving reliable interception of high-speed maneuvering UAV targets with higher overall efficiency in complex adversarial environments. Attached Figure Description

[0008] To more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0009] Figure 1 This is a schematic diagram of the architecture of a computer-aided laser source control system provided in one embodiment of the present invention; Figure 2 This is a flowchart illustrating a computer-aided laser source control method according to an embodiment of the present invention. Detailed Implementation

[0010] To further illustrate the technical means and effects adopted by the present invention to achieve its intended purpose, the specific implementation methods, structures, features, and effects of the present invention will be described in detail below with reference to the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0011] The following description, in conjunction with the accompanying drawings, details a specific scheme for a computer-aided laser source control method and system provided by the present invention.

[0012] For example, such as Figure 1 The diagram shown is a schematic representation of the architecture of a computer-aided laser source control system (hereinafter referred to as the source control system 10) according to an embodiment of the present invention. The source control system 10 includes: a prediction probability cloud module 11, a trajectory analysis module 12, an accuracy evaluation module 13, an expectation calculation module 14, and a source control module 15. The modules are described below in sequence: (1) Predictive probability cloud module 11.

[0013] The prediction probability cloud module 11 is responsible for acquiring and processing real-time perception data of the target UAV, generating a future position prediction distribution, i.e., the prediction probability cloud, to guide laser pre-aiming, providing a basic data framework for subsequent accurate assessment and aiming.

[0014] Optionally, the prediction probability cloud module 11 is used to acquire the current motion state data of the UAV and determine the prediction probability cloud based on the current motion state data. The prediction probability cloud contains multiple prediction locations, and each prediction location is associated with a probability cloud density predicted based on the current motion state of the UAV.

[0015] Specifically, the prediction probability cloud module 11 acquires real-time motion state data of the UAV, such as its three-dimensional coordinates, velocity, acceleration, and heading angle, using sensors integrated on the weapon station platform, including radar and electro-optical / infrared tracking systems. Then, within its internal processor, the prediction probability cloud module 11 initializes a series of particles (current spatial points) representing the UAV's possible current positions in the space surrounding the UAV based on this data. Next, the prediction probability cloud module 11 calls its stored preset motion model library (including models such as uniform velocity, uniform acceleration, and maneuvering turns) to match and assign a motion model that best fits the current state of each particle. Finally, the prediction probability cloud module 11 uses the assigned motion model to perform one or more extrapolations on each particle, calculating a series of predicted positions for the future interception period; then, by performing spatial density statistical analysis on these predicted positions, it assigns a probability cloud density value to each position, thus collectively forming a complete prediction probability cloud.

[0016] The prediction probability cloud module 11 outputs this prediction probability cloud to the trajectory analysis module 12 and the accuracy evaluation module 13 as common inputs for subsequent analysis.

[0017] (2) Trajectory Analysis Module 12.

[0018] The trajectory analysis module 12 is responsible for receiving UAV trajectory information from the sensor historical database. Through intelligent trajectory segmentation and pattern recognition technology, it extracts trajectory segments that can reflect the UAV's true attack intentions from the messy historical flight data, namely, error correction trajectory segments.

[0019] Optionally, the trajectory analysis module 12 is used to determine at least one error-correcting trajectory segment based on the historical trajectory data of the UAV. The error-correcting trajectory segment represents a segment of the UAV's trajectory after path replanning during its historical flight towards the attack target.

[0020] Specifically, the trajectory analysis module 12 first reads and caches a continuous historical trajectory point sequence of the UAV over a period of time. Then, the trajectory analysis module 12 analyzes the trajectory sequence, calculating a division coefficient for each trajectory point. This coefficient quantifies whether the point is in an "approximately stationary" state, similar to hovering or uniform linear motion. Based on a preset division threshold, the trajectory analysis module 12 identifies all division points, thereby automatically segmenting the continuous historical trajectory into multiple trajectory segments with independent motion patterns.

[0021] Furthermore, the trajectory analysis module 12 extracts the coordinate sequence of the starting part of each trajectory segment and, by analyzing its directional change characteristics (e.g., calculating the deviation of each point from the fitted line of the starting direction), initially filters out segments suspected of having undergone path correction (suspected error-correcting trajectories). Finally, the trajectory analysis module 12 determines the intersection points of the extension directions of all suspected error-correcting trajectories with the preset defense zone interface, performs three-dimensional spatial clustering analysis on all intersection points, and uses a density clustering algorithm to analyze these intersection points; the original historical trajectory segments corresponding to the intersection points belonging to the largest and densest clusters are ultimately determined to be error-correcting trajectory segments with a common directionality.

[0022] The trajectory analysis module 12 outputs these error-corrected trajectory segments to the accuracy evaluation module 13 as a key benchmark for evaluating the reliability of the predicted location.

[0023] (3) Accuracy assessment module 13.

[0024] The accuracy assessment module 13 is responsible for receiving the prediction probability cloud from the prediction probability cloud module 11 and the error correction trajectory segment from the trajectory analysis module 12. By calculating the matching degree between the motion direction of the predicted position and the historical intention direction, it assigns a target accuracy score to each predicted position in the prediction probability cloud, thereby quantifying the consistency between the predicted position and the actual attack target of the UAV.

[0025] Optionally, the accuracy evaluation module 13 is used to determine the target accuracy of the predicted position for each predicted position in the predicted probability cloud based on the degree of matching between the predicted position and the direction of motion of each error correction trajectory segment.

[0026] Specifically, for each predicted position in the probability cloud, the accuracy evaluation module 13 first calculates the motion direction vector pointing from the current time to the predicted position based on the particle motion model that generated the predicted position. Simultaneously, the accuracy evaluation module 13 extracts the overall trajectory direction vector for each error-corrected trajectory segment.

[0027] Next, the accuracy evaluation module 13 sequentially calculates the cosine similarity between the motion direction vector of the predicted location and the direction vector of each error correction trajectory segment, as a single direction similarity. Finally, the accuracy evaluation module 13 integrates all calculated direction similarities (e.g., calculates the arithmetic mean) and uses the resulting aggregated value as the target accuracy of the predicted location. The higher the target accuracy, the more the predicted location aligns with the historical attack intentions of the UAV.

[0028] The accuracy assessment module 13 outputs the accuracy of each predicted location and its target to the expectation calculation module 14.

[0029] (4) Expected calculation module 14.

[0030] The expected calculation module 14 is responsible for integrating multi-dimensional information such as target accuracy, spatial probability density, and target attitude to calculate the expected theoretical damage effect at each predicted position in the predicted probability cloud, providing a quantitative decision basis for the final selection of the best aiming point.

[0031] Optionally, the expectation calculation module 14 is used to determine the damage expectation at each predicted location in the predicted probability cloud based on the target accuracy, the corresponding probability cloud density, and the attitude information of the UAV at the predicted location.

[0032] Specifically, the expectation calculation module 14 receives target accuracy data from the accuracy evaluation module 13 and probability cloud density data corresponding to each predicted position from the prediction probability cloud module 11. Simultaneously, the expectation calculation module 14 integrates an attitude analysis submodule. This attitude analysis submodule, based on the UAV's 3D model and predicted position information, simulates and constructs a ray from the origin of the laser emitter through the centroid of the predicted probability cloud, and analyzes the intersection of this ray with the UAV's body at the predicted position, calculating relationships such as the ratio of its average equivalent cross-sectional area to the body volume, thereby determining an attitude damage factor. This factor characterizes the efficiency of laser energy transfer when the UAV is irradiated in this attitude.

[0033] Subsequently, the core processor of the expectation calculation module 14 normalizes and fuses the target accuracy, probability cloud density, and attitude damage factor (e.g., by multiplying the three and then standardizing), outputting a comprehensive damage expectation value. The predicted location with the highest damage expectation value is considered to be the point with the optimal combination of energy delivery efficiency and hit probability.

[0034] The expected calculation module 14 outputs a list of expected damage values ​​for all predicted locations to the light source control module 15.

[0035] (5) Light source control module 15.

[0036] The light source control module 15 is the system's execution terminal. It is responsible for accurately controlling the final aiming direction and emission power of the high-energy laser based on the decision results of the expectation calculation module 14, and implementing interception.

[0037] Optionally, the light source control module 15 is used to determine the aiming point from the predicted probability cloud based on the damage expectation of all predicted locations, and control the laser light source based on the aiming point.

[0038] Specifically, the light source control module 15 first receives a list of damage expectation values ​​from the expectation calculation module 14, and automatically selects the predicted position with the largest damage expectation value as the final aiming point. Then, the light source control module 15 drives the high-precision turntable to quickly align the laser emitter's beam axis with the spatial coordinates of the aiming point.

[0039] Simultaneously, the fire control calculation unit within the light source control module 15 calculates the minimum effective power required to achieve the desired damage effect based on real-time measured aiming point distance, current ambient temperature and humidity, visibility, and other environmental attenuation factors, combined with the laser's own energy characteristics. Finally, the light source control module 15 sends a trigger command to the laser power source (such as a supercapacitor bank), controlling it to release energy at the calculated emission power, generating a high-energy laser beam to continuously irradiate the aiming point, thereby completing the interception mission of the target UAV.

[0040] The above describes the computer-aided laser source control system 10 and its included modules.

[0041] For example, such as Figure 2 The diagram shown is a flowchart illustrating a computer-aided laser source control method according to an embodiment of the present invention, comprising the following steps: S201. Acquire the current motion state data of the UAV and determine the prediction probability cloud based on the current motion state data. The prediction probability cloud contains multiple prediction locations, and each prediction location is associated with a probability cloud density predicted based on the current motion state of the UAV.

[0042] For example, this step can be performed by the prediction probability cloud module 11 in the light source control system 10 described above. Specifically, when the prediction probability cloud module 11 acquires the current motion state data of the UAV, it can collect the target's coordinates, velocity, acceleration, and heading angle in real time through a multi-sensor fusion system integrated on the weapon platform.

[0043] Furthermore, the prediction probability cloud module 11 determines the prediction probability cloud based on the current motion state data, specifically including the following steps: (1) Based on the current motion state data, initialize multiple current space points in the space where the UAV is located, and select a matching motion model from the preset motion model library for each current space point.

[0044] Optionally, the prediction probability cloud module 11 generates a series of virtual particles in three-dimensional space according to a preset density rule, with the currently detected drone entity location as the center. These particles are the current spatial points.

[0045] Simultaneously, the prediction probability cloud module 11 calls its stored preset motion model library, which contains various typical motion models (such as uniform velocity, uniform acceleration, and cooperative turning models). The prediction probability cloud module 11 assigns a suitable motion model to each point by calculating the matching degree between the real-time motion information carried by each current spatial point and the parameters of each model. For example, the matching method is to calculate the likelihood function between the current state and the prediction residuals of each model.

[0046] (2) Predict the position of each current spatial point according to the selected motion model, obtain the corresponding predicted position, and determine the probability cloud density corresponding to each predicted position in the predicted probability cloud based on the distribution of all predicted positions.

[0047] Specifically, the probability cloud prediction module 11 uses the motion model assigned to each current spatial point to extrapolate and calculate the coordinates of that point after one or more future time steps, i.e., the predicted position. After performing the same operation on all current spatial points, a point cloud composed of a large number of predicted positions is obtained. The probability cloud prediction module 11 then uses a density estimation algorithm to analyze the point cloud, calculates the number of other predicted positions within a unit volume around each predicted position, and normalizes this density value to obtain the probability cloud density corresponding to that predicted position.

[0048] Thus, the prediction probability cloud module 11 generates a complete prediction probability cloud that includes the location and its probability of occurrence.

[0049] S202. Based on the historical trajectory data of the UAV, determine at least one error-correcting trajectory segment. The error-correcting trajectory segment represents a segment of the UAV's trajectory after path replanning during its historical flight towards the attack target.

[0050] For example, this step can be performed by the trajectory analysis module 12 in the light source control system 10 described above, and specifically includes the following steps: (1) Based on the historical trajectory data of the UAV, determine the dwell characteristics of the UAV on the historical trajectory, and divide the historical trajectory data into multiple historical trajectory segments according to the dwell characteristics; Optionally, when the trajectory analysis module 12 performs this step, it specifically includes: retrieving a continuous position sequence of the UAV over a past period from the trajectory database of the light source control system 10. For each trajectory point in the sequence, the trajectory analysis module 12 calculates its division coefficient, which is used to comprehensively characterize the probability that the point is in an approximately stationary state (such as hovering or uniform linear motion). The trajectory analysis module 12 presets a division threshold, marks all points with division coefficients greater than the threshold as division points, and then groups all trajectory points between adjacent division points into an independent historical trajectory segment. It should be noted that the specific process of dividing multiple historical trajectory segments is described in S301-S302 below, and will not be repeated here.

[0051] In another possible implementation, when the trajectory analysis module 12 divides multiple historical trajectory segments, it can also do so by identifying abrupt changes in the combination of speed and heading in the historical trajectory. For example, when the UAV simultaneously meets the conditions of speed below a threshold and heading change rate above a threshold, it is considered that it may have ended a maneuver and started a new segment, and this point can be used as the boundary of the trajectory segment.

[0052] Alternatively, the trajectory analysis module 12 can also employ a sliding window-based method for trajectory segment autoencoding and reconstruction error detection. The trajectory within the window is encoded and reconstructed; when the reconstruction error suddenly increases, it indicates a change in the motion pattern, thus determining the segmentation point.

[0053] Thus, the trajectory analysis module 12 lays the foundation for subsequent identification of trajectory segments with specific intentions by intelligently segmenting historical trajectories.

[0054] (2) Select the trajectory segments that meet the preset error correction characteristics from multiple historical trajectory segments as at least one error correction trajectory segment.

[0055] Optionally, when the trajectory analysis module 12 performs this step, it specifically includes: for each segmented historical trajectory segment, the trajectory analysis module 12 extracts its starting part and analyzes the directional deflection characteristics of this part of the trajectory. Specifically, it calculates the degree of deviation of the trajectory points relative to the baseline formed by the starting direction. When obvious directional adjustment characteristics are detected, the sub-trajectory after the adjustment is recorded as the suspected error correction trajectory of the trajectory segment. Then, the trajectory analysis module 12 determines the intersection points of the extension directions of all suspected error correction trajectories with the preset defense zone interface. Subsequently, the trajectory analysis module 12 uses a density-based clustering algorithm to analyze these landing points, identifying the cluster with the densest spatial location and the largest number of landing points as the "target pointing cluster". Finally, all historical trajectory segments whose intersection points with the defense zone interface belong to this "target pointing cluster" are determined as error correction trajectory segments. It should be noted that the specific process of dividing multiple historical trajectory segments is described in S303-S305 below, and will not be repeated here.

[0056] In another possible implementation, the trajectory analysis module 12 can also filter at least one error-correcting trajectory segment by analyzing the overall curvature characteristics of the trajectory segment. For example, only those trajectory segments with a small initial curvature (approximately a straight line) and an overall curvature below a certain threshold can be identified as error-correcting trajectory segments, since sharp turns are more likely to be obstacle avoidance than path correction.

[0057] Alternatively, when the trajectory analysis module 12 filters at least one error-correcting trajectory segment, it can also incorporate external geographic information. For example, trajectory segments whose initial heading avoids areas with known obstacles and points towards open airspace can be prioritized as error-correcting trajectory segments.

[0058] Therefore, the trajectory analysis module 12 extracts key trajectory segments that reflect the UAV's attack intent from the chaotic historical maneuvers through pattern recognition and cluster analysis.

[0059] S203. For each predicted location in the predicted probability cloud, determine the target accuracy of the predicted location based on the motion direction matching degree between the predicted location and each error correction trajectory segment. Target accuracy characterizes the degree of consistency between the predicted location and the UAV's attack target.

[0060] For example, this step can be performed by the accuracy evaluation module 13 in the light source control system 10 described above. Specifically, it includes: for any predicted position in the prediction probability cloud, the accuracy evaluation module 13 first calculates the motion direction vector from the current position of the UAV to the predicted position based on the motion model that generates the position. Simultaneously, for each error-correcting trajectory segment provided by the trajectory analysis module 12, the accuracy evaluation module 13 calculates the direction vector of its overall trajectory. Then, the accuracy evaluation module 13 calculates the directional similarity between the motion direction vector and each trajectory direction vector. Finally, the accuracy evaluation module 13 determines a comprehensive matching degree evaluation value based on all calculated directional similarities, as the target accuracy of the predicted position. It should be noted that the specific process for determining the target accuracy of the predicted position is described in S401-S403 below, and will not be repeated here.

[0061] In another possible implementation, when determining the target accuracy of the predicted location, the accuracy assessment module 13 can also introduce the "freshness" of the error correction trajectory segment as a weight based on the directional similarity. For example, the closer the error correction trajectory segment is to the current time, the higher the weight of its direction vector when calculating the overall matching degree, making the target accuracy more reflective of the latest intention of the UAV.

[0062] Alternatively, the accuracy assessment module 13 can also employ vector field consistency analysis. The direction vector of each error-correcting trajectory segment is treated as a directional guide within a local vector field, and the target accuracy is assessed by calculating the likelihood probability of the predicted position's motion direction within that vector field.

[0063] Therefore, the accuracy assessment module 13 assigns a confidence score to each predicted location through quantitative direction matching analysis. A high score means that the location is more likely to be on the path that the drone intends to fly to the target.

[0064] S204. For each predicted location in the predicted probability cloud, determine the expected damage at the predicted location based on the target accuracy, the corresponding probability cloud density, and the attitude information of the UAV at the predicted location.

[0065] For example, this step can be performed by the expectation calculation module 14 in the light source control system 10 described above, specifically including: the expectation calculation module 14 first determines the attitude damage factor based on the attitude information of the UAV at the predicted position. First, the expectation calculation module 14 analyzes the relationship between the equivalent irradiated cross section of the UAV at the predicted position and the laser beam when the laser beam is emitted from the emission source to the centroid of the predicted probability cloud. This relationship is used to calculate the attitude damage factor to characterize the energy receiving efficiency at this attitude. Then, the expectation calculation module 14 integrates the target accuracy, its probability cloud density, and the attitude damage factor at the predicted position, and calculates the damage expectation at the predicted position through a preset fusion calculation rule. It should be noted that the specific process for determining the damage expectation at the predicted position is described in S501-S502 below, and will not be repeated here.

[0066] In another possible implementation, when the expectation calculation module 14 determines the damage expectation at the predicted location, it can also introduce the energy attenuation rate of the laser as it travels through the atmosphere to that predicted location as an additional multiplicative factor into the calculation. This factor is calculated based on real-time atmospheric parameters and distance, making the damage expectation more realistically reflect the energy that ultimately reaches the target.

[0067] Alternatively, the expected calculation module 14 can also employ an expert system based on rule or case reasoning. This system stores historical damage effect assessments under different combinations of target accuracy, probability cloud density, and attitude. By matching the current parameter combination, it can infer the corresponding expected damage value.

[0068] Therefore, the expectation calculation module 14 calculates a comprehensive damage expectation index for optimal selection by integrating the hit probability, spatial distribution probability and damage physical effectiveness.

[0069] S205. Based on the damage expectation of all predicted locations, determine the aiming point from the predicted probability cloud, and control the laser source based on the aiming point.

[0070] For example, this step can be performed by the light source control module 15 in the light source control system 10 described above, and specifically includes the following steps: (1) Determine the emission power of the laser source based on the distance between the aiming point and the laser source and the environmental attenuation factors.

[0071] Specifically, the light source control module 15 receives a damage expectation list from the expectation calculation module 14 and automatically selects the predicted position with the highest damage expectation value as the final aiming point. The fire control calculation unit within the light source control module 15 obtains the aiming point distance based on real-time ranging information and queries the current atmospheric database to obtain the attenuation coefficient. Based on the distance, attenuation coefficient, and target energy density required to achieve damage, this unit calculates the minimum effective power required to emit the laser using a physical model.

[0072] (2) Control the laser source to illuminate the aiming point with emission power.

[0073] Furthermore, the light source control module 15 sends commands to the high-precision servo turntable, driving the laser emitter beam axis to precisely align with the spatial coordinates of the aiming point. Simultaneously, the light source control module 15 sends power loading commands to the laser power source, controlling it to release the calculated emission power, forming a high-energy laser beam to continuously irradiate the aiming point until the expected energy accumulation effect is achieved or the target is destroyed.

[0074] Based on the above technical solution, this invention integrates real-time UAV motion prediction, historical intent trajectory analysis, and damage physics modeling to achieve intelligent selection of laser aiming points and precise power control. Its technical effects are as follows: First, by analyzing "error correction" segments in historical trajectories, the true attack intent direction of the UAV is extracted, thereby correcting the reliability of the probability cloud based solely on instantaneous state predictions, significantly improving the consistency between the predicted position and the actual target direction. Second, by introducing an attitude damage factor, the physical attitude of the UAV when irradiated is incorporated into the evaluation system, ensuring that the final selected aiming point is not only a high-probability hit point but also a high-energy-efficiency damage point. Finally, based on comprehensive damage expectations, the aiming point is determined and the required power is dynamically calculated, achieving reliable interception of high-speed maneuvering UAV targets with higher overall efficiency in complex adversarial environments.

[0075] In one possible implementation, an embodiment of the present invention provides another computer-aided laser source control method, which determines at least one error-correction trajectory segment based on the historical trajectory data of the UAV, specifically including the following steps: S301. For each trajectory point in the historical trajectory, calculate the corresponding division coefficient. The division coefficient is used to characterize the motion state type of the trajectory point.

[0076] It should be noted that each trajectory point in the historical trajectory refers to a continuous time-series location point obtained from the sensor at fixed sampling time intervals.

[0077] Optionally, when performing this step, the trajectory analysis module 12 first obtains the hovering coefficient of the i-th trajectory point required for calculation. The dimensionless distance to the i-th trajectory point : First, starting from the current i-th trajectory point, take L consecutive trajectory points backward (e.g., L=5). Count the number of these L points whose position coordinates (3D coordinates) have changed compared to their previous point, denoted as c. Then the hovering coefficient... If the positions of all points remain unchanged, then c = 0. =0 indicates that it may be in a hovering state.

[0078] Secondly, starting from the i-th trajectory point, take L consecutive trajectory points forward. Using the spatial coordinates of these L points, fit a three-dimensional straight line through linear regression. Then, calculate the Euclidean distance from each of these L points to the fitted line, and sum all the distances to obtain the distance sum. Calculate the Euclidean distance between the first and last points of these L trajectory points. Then combine the distance and Divide by Obtain the dimensionless distance and . The smaller the value, the closer the spatial distribution of these L points is to a straight line.

[0079] Furthermore, for the i-th trajectory point in the historical trajectory data sequence, its partitioning coefficient is calculated using the following formula. :

[0080] in, This represents the partitioning coefficient of the i-th trajectory point; This represents the hovering coefficient of the i-th trajectory point; This represents the dimensionless sum of distances corresponding to the i-th trajectory point; This represents an adjustment factor, which is a very small positive value, such as 10 to the power of negative 5, used to ensure that the formula... Numerical stability as it approaches 0.

[0081] It should be noted that the above formula is obtained by taking... and The maximum value is used to achieve hovering state ( Small) and uniform linear motion ( The "or" logical judgment (small). When the drone is hovering, Close to 0, but It is also close to 0, making It's very big. The value is relatively large; when the drone is moving at a constant velocity in a straight line... Close to 1, but Approaching 0 also makes It's very big. The value is relatively large; when the drone performs irregular maneuvers, Close to 1 and Larger Smaller The value is approximately equal to 1. By setting a reasonable threshold (e.g., 10), it is possible to filter out... Trajectory points exceeding a threshold are used as dividing points to accurately identify "approximately stationary" states.

[0082] S302. Identify trajectory points with a division coefficient greater than a preset division threshold as division points, and divide the historical trajectory into multiple historical trajectory segments based on the trajectory points between adjacent division points.

[0083] Furthermore, the trajectory analysis module 12 presets a segmentation threshold. The trajectory analysis module 12 will then divide all the segments calculated in step S301... Compare with the partitioning threshold and select all that meet the criteria. Trajectory points that exceed the division threshold are marked as division points.

[0084] For example, the above-mentioned threshold value can be 0.75. The specific value selection rule can be: based on statistical analysis of a large amount of historical trajectory data, the threshold value for the vast majority of "approximately stationary" points (such as hovering, uniform linear motion) is determined. It can exceed this threshold, while the normal maneuver point... The value is below this threshold. It should be noted that, although... No normalization was performed, but because After dimensionless processing (eliminating the influence of dimensions to make distances comparable), and since the threshold is obtained based on statistical learning from historical data, therefore... The distribution in actual data is relatively stable, and the threshold comparison logic is feasible and effective in engineering practice.

[0085] Next, the trajectory analysis module 12 traverses all trajectory points in chronological order, grouping all trajectory points (including the start and end points) between every two adjacent dividing points to form an independent historical trajectory segment. If multiple dividing points are consecutively adjacent, only the last dividing point is used as the valid segment boundary. Through this operation, the complete historical trajectory data is divided into multiple historical trajectory segments that are relatively independent in terms of motion patterns.

[0086] S303. For each historical trajectory segment, analyze the directional change characteristics of the initial part of the trajectory to obtain the suspected error correction trajectory.

[0087] In this step, the trajectory analysis module 12 analyzes the starting part (e.g., the first b×) of the k-th historical trajectory segment obtained from the division. There are trajectory points, where b is a preset scaling factor. Perform a directional consistency analysis on the total length of the trajectory segment. For the t-th trajectory point in this segment, calculate its directional deflection using the following formula. : in, This represents the degree of directional deflection of the t-th trajectory point within the initial portion of the k-th historical trajectory segment. This represents the spatial coordinates of the trajectory point; This represents a straight line fitted to the direction of the vector formed by all points in the time series from the starting point (the 1st point) to the tth point of the trajectory segment. Point to the straight line The Euclidean distance; This represents a preset distance normalization constant (e.g., 10 meters) used to make the input dimensionless; This represents the hyperbolic tangent function.

[0088] It should be noted that the above formula calculates the trajectory points. To the initial overall trend of its historical trajectory segment Euclidean distance This is used to quantify the degree of spatial deviation at a point, thus reflecting the severity of the directional adjustment. To transform this dimensional distance into a dimensionless scalar suitable for threshold judgment, the formula uses a hyperbolic tangent function for mapping and introduces a preset distance normalization constant. (For example, 10 meters), making the input parameter dimensionless. The hyperbolic tangent function has an output range of [0, 1) when the input is non-negative: when the distance is 0, the output is 0, indicating no directional deflection; as the distance increases, the output asymptotically approaches 1, which more reasonably reflects the increase in the degree of deflection. This leads to... The closer the value is to 1, the more significant the directional correction occurred at that trajectory point.

[0089] Furthermore, the trajectory analysis module 12 calculates the degree of directional deviation of each trajectory point within the initial part of the current historical trajectory segment in chronological order, starting from the starting point of the current historical trajectory segment. t. When a certain trajectory point first appears. If the deviation exceeds a preset directional deflection threshold, it is determined that a significant directional correction has occurred in the sub-trajectory between the starting point of the historical trajectory segment and the trajectory point t. The trajectory analysis module 12 then marks this sub-trajectory segment as a suspected error-correction trajectory of the historical trajectory segment.

[0090] For example, the aforementioned orientation deflection threshold can be set to 0.7. The specific rule for setting this threshold is: based on statistical analysis of historically validated error correction trajectory samples, to ensure reliable capture of obvious initial orientation adjustments made to realign the target, while filtering out unintentional minor attitude fluctuations or measurement noise. In practical applications, the orientation deflection threshold needs to be reset based on the output characteristics of the hyperbolic tangent function (e.g., set between 0.7 and 0.8) and calibrated experimentally.

[0091] S304. Determine the intersection point between the extension direction of each suspected error correction trajectory and the preset defense zone interface, and perform three-dimensional spatial cluster analysis on all intersection points.

[0092] It should be noted that the preset defense zone interface refers to a virtual three-dimensional boundary surface (such as a hemispherical, cylindrical, or planar composite interface) set to protect a specific target, and its geometric parameters are preset according to the protection scenario. By extending the trajectory to this interface and finding the intersection, the directional information of the trajectory in three-dimensional space can be preserved, avoiding the loss of height dimension information due to projection onto a two-dimensional plane.

[0093] Specifically, for each historical trajectory segment, the trajectory analysis module 12 extracts the overall motion direction vector of the suspected error correction trajectory. Using this direction vector as the extension direction, the trajectory analysis module 12 calculates the intersection point with the preset defense zone interface (e.g., a hemisphere or cylinder with radius R centered on the protected target), thus obtaining the defense zone interface intersection point (three-dimensional spatial point) corresponding to the trajectory segment.

[0094] Furthermore, after obtaining the intersection points of the defense zone interfaces corresponding to all historical trajectory segments, the trajectory analysis module 12 uses a density-based three-dimensional spatial clustering algorithm (such as the three-dimensional DBSCAN algorithm) to perform cluster analysis on all intersection points. This algorithm can automatically group intersection points with similar positions into the same cluster and identify outliers based on the density distribution of intersection points in three-dimensional space.

[0095] S305. The historical trajectory segment corresponding to the intersection of the defense zone interface belonging to the largest cluster is determined as the error correction trajectory segment.

[0096] For example, after completing the cluster analysis, the trajectory analysis module 12 counts the number of intersections in each three-dimensional cluster. The trajectory analysis module 12 identifies the cluster with the most intersections as the largest cluster. The physical meaning is that the area with the most intersections likely corresponds to the common pointing direction of the UAV attack targets on the defense zone interface.

[0097] Finally, the trajectory analysis module 12 identifies all historical trajectory segments whose intersections at all defense zone interfaces belong to the largest cluster as error-correcting trajectory segments. These error-correcting trajectory segments are considered to be trajectory fragments with a common intention generated by the UAV during its historical flight as it replans its path toward the attack target.

[0098] Based on the above technical solution, this embodiment of the invention intelligently segments historical trajectories by calculating quantitative division coefficients, and accurately filters out error-correcting trajectory segments that reflect the true attack intent of the UAV from complex historical flight data by analyzing the directional deflection characteristics of the starting part of the trajectory segment and the spatial clustering of the intersection point of the defense zone interface. This overcomes the limitations of relying solely on instantaneous states and provides crucial historical intent information for subsequently accurately assessing the target accuracy of the predicted location.

[0099] In one possible implementation, an embodiment of the present invention provides another computer-aided laser source control method, in which the target accuracy of the predicted position is determined for each predicted position in the predicted probability cloud based on the motion direction matching degree between the predicted position and each error correction trajectory segment, specifically including the following steps: S401. Obtain the motion direction vector of the predicted position and the trajectory direction vector of each error correction trajectory segment.

[0100] In this step, the accuracy evaluation module 13 performs the following operations for any predicted position (denoted as the x-th predicted position) in the predicted probability cloud: First, based on the current spatial point and its motion model on which the predicted position is generated, it calculates the vector pointing from the current spatial point (corresponding to the estimated position of the UAV at the current moment) to the x-th predicted position, and defines this vector as the motion direction vector of the predicted position. .

[0101] Simultaneously, the accuracy evaluation module 13 iterates through all N error-correcting trajectory segments provided by the trajectory analysis module 12. For the nth error-correcting trajectory segment, the accuracy evaluation module 13 extracts the principal direction of the spatial distribution of all points in the trajectory segment through principal component analysis, and defines this principal direction vector as the trajectory direction vector of the error-correcting trajectory segment. .

[0102] S402. Calculate the directional similarity between the motion direction vector and each trajectory direction vector.

[0103] For example, the accuracy evaluation module 13 calculates the directional similarity between the motion direction vector and each trajectory direction vector according to the following formula:

[0104] in, This represents the directional similarity between the motion direction vector of the x-th predicted position and the trajectory direction vector of the n-th error-corrected trajectory segment; This represents the motion direction vector at the x-th predicted position; This represents the trajectory direction vector of the nth error-correcting trajectory segment; This represents the cosine similarity function, used to calculate the cosine value of the angle between two vectors.

[0105] Understandably, the cosine similarity function By calculating the ratio of the dot product of two vectors to the product of their respective magnitudes, a scalar value between -1 and 1 is obtained. The closer this value is to 1, the stronger the vector... and The closer the direction is to -1, the more opposite the direction; the closer to 0, the closer the direction is to orthogonal. To eliminate the interference of negative values ​​on subsequent calculations, a linear transformation is used. Map it to the interval [0, 1]. Transformed The closer the value is to 1, the more consistent the direction of movement of the x-th predicted position is with the historical intention direction reflected by the n-th error correction trajectory segment; the closer the value is to 0, the more inconsistent or opposite the directions are.

[0106] S403. Determine the target accuracy of the predicted location based on the similarity in all directions.

[0107] Furthermore, the accuracy assessment module 13 calculates the directional similarity between the motion direction vector of the x-th predicted position and all N error-correcting trajectory segments based on the aforementioned steps. arrive Then, the target accuracy of the predicted location is calculated using the following formula:

[0108] in, This represents the target accuracy at the x-th predicted position; N represents the total number of error-correcting trajectory segments. This represents the directional similarity between the x-th predicted position and the n-th error-corrected trajectory segment.

[0109] It should be noted that the target accuracy in the above formula Similarity in all directions The arithmetic mean of the predicted location. Its physical meaning is that by averaging the consistency between the predicted location's direction of motion and the directions of all historical error correction trajectories, it comprehensively assesses the overall degree of consistency between the predicted location and the drone's historical attack intentions. The higher the value, the more consistent the predicted direction of movement is with the direction toward the target after multiple path replannings in the past, on average. Therefore, the higher the probability (from the perspective of intent) that the drone will appear at the predicted location in the future, the higher its credibility or accuracy as a targeting point.

[0110] Based on the above technical solution, this embodiment of the invention calculates and averages the cosine similarity between the predicted location's motion direction and the error-correcting trajectory direction reflecting historical attack intentions, assigning a quantified target accuracy score to each predicted location. This method effectively utilizes the target pointing information contained in historical trajectories, compensating for the shortcomings of relying solely on the current instantaneous state for prediction. This allows for more accurate identification of predicted locations that better match the actual attack intentions of the drone, significantly improving the logical rationality and expected hit probability of laser aiming point selection.

[0111] In one possible implementation, an embodiment of the present invention provides another computer-aided laser source control method, in which, for each predicted location in the predicted probability cloud, the expected damage at the predicted location is determined based on the target accuracy, the corresponding probability cloud density, and the attitude information of the UAV at the predicted location. Specifically, this includes the following steps: S501. Determine the attitude damage factor based on the attitude information of the UAV at the predicted location. The attitude damage factor characterizes the degree to which the UAV's attitude affects the laser damage effect.

[0112] For example, when the expected calculation module 14 performs this step, it specifically includes the following steps: (1) Construct a ray that passes through the centroid of the predicted probability cloud and originates from a laser source.

[0113] In a specific example, the expected calculation module 14 first calculates the arithmetic mean of the coordinates of all predicted positions in the entire predicted probability cloud and defines it as the centroid of the predicted probability cloud. Then, taking the emission point of the laser source (the origin of the system coordinate system) as the starting point and the direction pointing to the centroid as the direction, a ray is defined in three-dimensional space.

[0114] (2) Determine the correlation between the ray and the average equivalent cross-sectional area of ​​the UAV at the predicted location, and determine the attitude damage factor based on the correlation.

[0115] Furthermore, the expectation calculation module 14 performs the following operations for the x-th predicted position in the predicted probability cloud: taking the ray direction as a reference, based on the average effective irradiated area of ​​the laser in that direction under the typical attitude distribution of the UAV. This area serves as the equivalent cross-sectional area. It can be statistically estimated using the angle between the UAV's geometric model and the ray direction, reflecting the typical effective area of ​​the UAV illuminated at this relative orientation. Then, the expectation calculation module 14 calculates the attitude damage factor for the x-th predicted position. ,in This represents the maximum effective illuminated area achievable by the UAV in all possible attitudes (a constant calculated beforehand based on the UAV's geometric model). It can be understood that the attitude damage factor is the ratio of the current average effective illuminated area to the maximum possible illuminated area. This ratio... The effective area ratio of the drone in this posture relative to the optimal illuminated posture was quantified. The larger the value (closer to 1), the larger the effective exposed area of ​​the drone to the laser beam when it is irradiated in this posture, meaning that this posture is more favorable for laser damage.

[0116] S502. Determine the damage expectation at the predicted location based on the target accuracy at the predicted location, the probability cloud density corresponding to the predicted location, and the attitude damage factor.

[0117] For example, the expectation calculation module 14 calculates the expected damage at the x-th predicted location using the following formula. :

[0118] in, This indicates the accuracy of the target corresponding to the x-th predicted position; This represents the probability cloud density corresponding to the x-th predicted location; Represents the attitude damage factor at the x-th predicted position; This represents the maximum-minimum normalization function, where the maximum and minimum values ​​are preset empirical extreme values ​​obtained based on a large amount of historical experimental data. If the calculation result exceeds the interval [0,1], it is restricted to the range [0,1] by a truncation function (i.e., if the result is less than 0, it is taken as 0, and if it is greater than 1, it is taken as 1) to eliminate the influence of outliers on the evaluation index.

[0119] It should be noted that the expected damage in the above formula... The calculation incorporates target accuracy (Reflecting the likelihood of hitting the intended target), probability cloud density (Reflecting the probability of occurrence based on kinematic predictions) and posture damage factors (Reflecting the physical efficiency of damage) Three aspects of information. Formula First, the first two factors reflecting the probability of a hit are multiplied and normalized to obtain a comprehensive weighted probability of "hitting and appearing here"; then, this weighted probability is multiplied by the attitude damage factor. Multiply by each product to obtain the final expected damage value. Here It has dimensions (reciprocal of length), and its physical meaning is the ratio of the effective exposed area of ​​a unit volume UAV to the laser beam under the predicted position and attitude. The larger the value, the larger the area receiving laser energy under the same volume, and the higher the damage potential. Although It has dimensions, but when comparing different predicted locations... When the value is given, since the same volume V is used at all locations, and The calculation method is consistent, therefore The relative size depends only on the product of the probability weighting at each location and the effective exposure area; the dimensions do not affect the comparison results and can reasonably reflect the damage effectiveness. Therefore, The value reflects the combined expectations of three dimensions: whether the laser beam can hit the point, whether the drone can fly to the point, and whether the damage effect after hitting the point is good enough.

[0120] Based on the above technical solution, this invention quantifies the attitude damage factor by constructing a ray and analyzing the average equivalent cross-sectional area of ​​the UAV at the predicted position. This factor is then fused with target accuracy and probability cloud density to calculate the damage expectation. This method combines the accuracy of trajectory prediction with damage physics effectiveness, ensuring that the selection of the aiming point not only pursues high hit probability but also high damage efficiency, thereby significantly improving the overall interception effectiveness of laser weapons against high-speed maneuvering targets under energy constraints.

[0121] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0122] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

Claims

1. A computer-aided laser source control method, characterized in that, The method includes: The current motion state data of the UAV is acquired, and a predicted probability cloud is determined based on the current motion state data; wherein the predicted probability cloud contains multiple predicted locations, and each predicted location is associated with a probability cloud density predicted based on the current motion state of the UAV. Based on the historical trajectory data of the UAV, at least one error correction trajectory segment is determined; wherein, the error correction trajectory segment is used to characterize the trajectory segment of the UAV after path replanning in order to move toward the attack target during historical flight; For each predicted location in the predicted probability cloud, the target accuracy of the predicted location is determined based on the motion direction matching degree between the predicted location and each error correction trajectory segment; wherein, the target accuracy is used to characterize the degree of consistency between the predicted location and the attack target of the UAV; For each predicted location in the predicted probability cloud, the expected damage at the predicted location is determined based on the target accuracy, the corresponding probability cloud density, and the attitude information of the UAV at the predicted location. Based on the damage expectations at all predicted locations, an aiming point is determined from the predicted probability cloud, and the laser source is controlled based on the aiming point.

2. The computer-aided laser source control method according to claim 1, characterized in that, Determining the predicted probability cloud based on the current motion state data specifically includes: Based on the current motion state data, multiple current spatial points are initialized in the space where the UAV is located, and a matching motion model is selected from a preset motion model library for each current spatial point; Based on the selected motion model, the position of each current spatial point is predicted to obtain the corresponding predicted position. Based on the distribution of all predicted positions, the probability cloud density corresponding to each predicted position in the predicted probability cloud is determined.

3. The computer-aided laser source control method according to claim 1, characterized in that, Based on the historical trajectory data of the UAV, at least one error correction trajectory segment is determined, specifically including: Based on the historical trajectory data of the UAV, the dwell characteristics of the UAV on the historical trajectory are determined, and the historical trajectory data is divided into multiple historical trajectory segments based on the dwell characteristics; From the multiple historical trajectory segments, trajectory segments that meet the preset error correction characteristics are selected as at least one error correction trajectory segment.

4. The computer-aided laser source control method according to claim 3, characterized in that, The historical trajectory data is divided into multiple historical trajectory segments based on the dwelling characteristics, specifically including: For each trajectory point in the historical trajectory, a corresponding division coefficient is calculated; wherein, the division coefficient is used to characterize the motion state type of the trajectory point; Trajectory points with a division coefficient greater than a preset division threshold are identified as division points, and the historical trajectory is divided into multiple historical trajectory segments based on the trajectory points between adjacent division points.

5. The computer-aided laser source control method according to claim 3, characterized in that, From the multiple historical trajectory segments, trajectory segments that meet the preset error correction characteristics are selected, specifically including: For each historical trajectory segment, the directional change characteristics of the initial part of the trajectory are analyzed to obtain the suspected error correction trajectory; Determine the intersection point between the extension direction of each suspected error correction trajectory and the preset defense zone interface, and perform three-dimensional spatial cluster analysis on all intersection points; The historical trajectory segment corresponding to the intersection of the defense zone interface belonging to the largest cluster is determined as the error correction trajectory segment.

6. The computer-aided laser source control method according to claim 1, characterized in that, For each predicted location in the predicted probability cloud, the target accuracy of the predicted location is determined based on the motion direction matching degree between the predicted location and each error correction trajectory segment, specifically including: Obtain the motion direction vector of the predicted position and the trajectory direction vector of each error correction trajectory segment, respectively; Calculate the directional similarity between the motion direction vector and each trajectory direction vector; The target accuracy of the predicted location is determined based on the similarity in all directions.

7. The computer-aided laser source control method according to claim 1, characterized in that, For each predicted location in the predicted probability cloud, based on the target accuracy, the corresponding probability cloud density, and the attitude information of the UAV at the predicted location, the expected damage at the predicted location is determined, specifically including: Based on the attitude information of the UAV at the predicted location, an attitude damage factor is determined; wherein, the attitude damage factor is used to characterize the degree of influence of the UAV attitude on the laser damage effect; The damage expectation at the predicted location is determined based on the target accuracy at the predicted location, the probability cloud density corresponding to the predicted location, and the attitude damage factor.

8. The computer-aided laser source control method according to claim 7, characterized in that, Based on the attitude information of the UAV at the predicted location, the attitude damage factor is determined, specifically including: Construct a ray that passes through the centroid of the predicted probability cloud and originates from a laser source; The correlation between the ray and the average equivalent cross-sectional area of ​​the UAV at the predicted location is determined, and the attitude damage factor is determined based on the correlation.

9. The computer-aided laser source control method according to claim 1, characterized in that, Controlling the laser source based on the aiming point specifically includes: The emission power of the laser source is determined based on the distance between the aiming point and the laser source, as well as environmental attenuation factors. The laser source is controlled to illuminate the aiming point at the emission power.

10. A computer-aided laser source control system, characterized in that, The system includes: a prediction probability cloud module, a trajectory analysis module, an accuracy evaluation module, an expectation calculation module, and a light source control module; The prediction probability cloud module is used to acquire the current motion state data of the UAV and determine the prediction probability cloud based on the current motion state data; wherein, the prediction probability cloud contains multiple prediction locations, and each prediction location is associated with a probability cloud density predicted based on the current motion state of the UAV. The trajectory analysis module is used to determine at least one error-correcting trajectory segment based on the historical trajectory data of the UAV; wherein, the error-correcting trajectory segment is used to characterize the trajectory segment of the UAV after path replanning in order to move toward the attack target during historical flight; The accuracy evaluation module is used to determine the target accuracy of each predicted position in the predicted probability cloud based on the motion direction matching degree between the predicted position and each error correction trajectory segment; wherein, the target accuracy is used to characterize the degree of consistency between the predicted position and the attack target of the UAV. The expectation calculation module is used to determine the expected damage at each predicted location in the predicted probability cloud, based on the target accuracy, the corresponding probability cloud density, and the attitude information of the UAV at the predicted location. The light source control module is used to determine the aiming point from the predicted probability cloud based on the damage expectation of all predicted locations, and control the laser light source based on the aiming point.

Images