Air-drop target prediction method, device and equipment for multi-dimensional parameter dynamic evolution under multi-model driving

Abstract

The invention relates to an airdrop target prediction method, device and equipment for multi-dimensional parameter dynamic evolution under multi-model driving. The method comprises a multi-model-driven multi-dimensional parameter dynamic evolution air-drop target prediction method, device and equipment, under a drag force model of an object-parachute system, the relation between the parachute unfolding area and the resistance in the parachute opening process of the object-parachute system is considered, an object-parachute system motion model is constructed, and the object-parachute system motion model is used for predicting the object-parachute in a multi-model-driven multi-dimensional parameter dynamic evolution air-drop target. Obtaining wind field information of different height layers according to the obtained air-drop height, calculating an air-drop trajectory to obtain a predicted landing site based on an object-parachute system motion model according to the air-drop initial information and the wind field information of the different height layers, and correcting the air-drop initial information according to a target landing site and the predicted landing site to obtain a target landing site; and carrying out air-drop according to the corrected air-drop information. By adopting the method, accurate air-drop can be performed according to the target landing site.

Description

Technical Field

[0001] The present application relates to the field of airdrop technology, and in particular to a method, device and equipment for predicting airdrop targets with dynamic evolution of multi-dimensional parameters driven by multiple models. Background Art

[0002] In the civilian sector, the importance of precision airdrop technology is becoming increasingly prominent. In disaster relief scenarios, the rapid and accurate delivery of relief supplies to affected areas is crucial. For example, airdrop systems designed for disaster relief, with precise landing capabilities, can meet the needs of special disaster relief situations, providing timely access to essential supplies for disaster victims, significantly improving rescue efficiency, saving lives, and minimizing losses. In the medical rescue sector, airdrop systems for emergency medical rescue can adapt to the airdrop requirements of specific aircraft, making emergency medical assistance possible in remote or inaccessible areas. In logistics and supply chain management, the introduction of precision airdrop technology can achieve accurate and rapid distribution, improving logistics efficiency and reducing costs.

[0003] Currently, wind field detection in precision airdrop technology primarily relies on weather balloons and radar. Balloon wind measurement uses changes in the trajectory of a balloon to detect wind field distribution. Because balloon wind measurement shares similar aerodynamic characteristics with parachutes, the wind field measurements are more suitable for parachute airdrop calculations. Wind radar, on the other hand, uses the scattering of electromagnetic waves by atmospheric turbulence or the velocity of aerosols in the air to detect wind fields. It offers advantages such as high temporal and spatial resolution, strong real-time capabilities, and all-weather detection.

[0004] However, there are still many problems with the existing technology. In terms of wind field detection, weather balloon wind measurement has the problems of long detection cycle and poor timeliness. It is difficult to provide accurate wind field information in an emergency situation and cannot cope with relatively harsh wind field environments. In terms of airdrop target prediction methods, current research focuses on the motion deviation of the parachute system caused by wind. Although some algorithms take into account the impact of wind on the parachute, they have limitations. For example, some algorithms only rely on the theoretical model of the wind field distribution with height in a certain area for simulation analysis, which cannot quantitatively explain the performance and accuracy of the algorithm, and lacks measured control data for feasibility verification; some algorithms do not consider the impact of the external wind field on the parachute motion state when establishing the model, resulting in inaccurate prediction of airdrop targets, which makes it difficult to meet the strict requirements for precise airdrops in practical applications. Summary of the Invention

[0005] Based on this, it is necessary to provide an airdrop target prediction method, device and equipment with dynamic evolution of multi-dimensional parameters driven by multiple models that can achieve precise airdrops in response to the above technical problems.

[0006] A method for airdrop target prediction based on dynamic evolution of multi-dimensional parameters driven by multiple models, the method comprising: Under the drag model of the cargo-parachute system, the relationship between the parachute deployment area and the drag during the parachute opening process is considered, and the motion model of the cargo-parachute system is constructed. Obtaining the airdrop height, initial airdrop information, and target landing location, and obtaining wind field information at different altitudes based on the airdrop height; Based on the cargo-parachute system motion model, the airdrop trajectory is calculated according to the initial airdrop information and wind field information at different altitudes to obtain a predicted landing location; The initial airdrop information is corrected according to the target landing location and the predicted landing location, and the airdrop is performed according to the corrected airdrop information.

[0007] In one embodiment, the drag model of the cargo-parachute system is expressed as: ; In the above formula, is the wing drag coefficient, Indicates the air density at the height of the parachute system. represents the equivalent windward diameter of the parachute system, is the background ground wind speed, is the velocity of the cargo-parachute system over the ground.

[0008] In one embodiment, the object-parachute system motion model is expressed as: ; In the above formula, 、 、 They represent the gravitational acceleration of the parachute system under the force in the X, Y, and Z directions respectively. 、 、 are the sum of wind speed and parachute speed in X, Y and Z directions respectively, represents the gravity of the object-parachute system, Indicates the quality of the parachute system.

[0009] In one embodiment, when the airdrop trajectory is calculated based on the cargo-parachute system motion model, the initial airdrop information, and the wind field information at different altitudes to obtain the predicted landing location: Based on the cargo-parachute system motion model, the velocity and displacement at the next moment are calculated according to the initial airdrop information and the wind field information of the corresponding altitude layer to obtain the position coordinates at the next moment; According to the position coordinates at the current moment, the position coordinates at the next moment are calculated in sequence using the object-parachute system motion model to obtain the airdrop trajectory.

[0010] In one embodiment, when calculating the airdrop trajectory, the position coordinates at a certain moment are expressed as: ; In the above formula, Indicates the initial position coordinates, or the current position coordinates, Indicates the initial speed, or current speed, Indicates the next moment, represents the acceleration obtained based on the object-parachute system motion model, Represents the integral variable, which is used to integrate the acceleration in the time domain.

[0011] In one embodiment, the Runge-Kutta method is used to calculate the airdrop trajectory.

[0012] In one embodiment, the correcting the airdrop initial information according to the target landing location and the predicted landing location includes: According to the target landing location and the predicted landing location, a left-right drift distance and a front-back drift distance are obtained; According to the left-right drift distance and the front-back drift distance, the initial airdrop speed of the initial airdrop information and the parachute opening time in the cargo-parachute system are corrected.

[0013] The present application also provides an airdrop target prediction device with dynamic evolution of multi-dimensional parameters driven by multiple models, the device comprising: The cargo-parachute system motion model construction module is used to consider the relationship between the parachute deployment area and the drag during the parachute opening process under the drag model of the cargo-parachute system, and to construct the cargo-parachute system motion model; An information acquisition module is used to obtain the airdrop height, initial airdrop information, and target landing location, and obtain wind field information at different altitudes based on the airdrop height; A landing location prediction module is used to calculate the airdrop trajectory based on the cargo-parachute system motion model, the initial airdrop information, and wind field information at different altitudes to obtain a predicted landing location; The airdrop information correction module is used to correct the initial airdrop information according to the target landing location and the predicted landing location, and perform airdrop according to the corrected airdrop information.

[0014] A computer device includes a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented: Under the drag model of the cargo-parachute system, the relationship between the parachute deployment area and the drag during the parachute opening process is considered, and the motion model of the cargo-parachute system is constructed. Obtaining the airdrop height, initial airdrop information, and target landing location, and obtaining wind field information at different altitudes based on the airdrop height; Based on the cargo-parachute system motion model, the airdrop trajectory is calculated according to the initial airdrop information and wind field information at different altitudes to obtain a predicted landing location; The initial airdrop information is corrected according to the target landing location and the predicted landing location, and the airdrop is performed according to the corrected airdrop information.

[0015] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the following steps: Under the drag model of the cargo-parachute system, the relationship between the parachute deployment area and the drag during the parachute opening process is considered, and the motion model of the cargo-parachute system is constructed. Obtaining the airdrop height, initial airdrop information, and target landing location, and obtaining wind field information at different altitudes based on the airdrop height; Based on the cargo-parachute system motion model, the airdrop trajectory is calculated according to the initial airdrop information and wind field information at different altitudes to obtain a predicted landing location; The initial airdrop information is corrected according to the target landing location and the predicted landing location, and the airdrop is performed according to the corrected airdrop information.

[0016] The aforementioned multi-model-driven airdrop target prediction method, device, and equipment with dynamic evolution of multi-dimensional parameters constructs a cargo-parachute system motion model by considering the relationship between the parachute's deployed area and drag during the deployment process within the cargo-parachute system's drag model. Wind field information at different altitudes is obtained based on the acquired airdrop altitude. Based on the cargo-parachute system motion model, the airdrop trajectory is calculated based on the initial airdrop information and wind field information at different altitudes to obtain a predicted landing location. The initial airdrop information is corrected based on the target and predicted landing locations, and the airdrop is performed according to the corrected airdrop information. This method enables precise airdrops based on the target landing location. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 1 is a flow chart of an airdrop target prediction method with dynamic evolution of multi-dimensional parameters driven by multiple models in one embodiment; Figure 2 A schematic diagram of a force analysis of a parachute system in one embodiment; Figure 3 FIG. 1 is a schematic diagram of a simulation of a calculation process using a first-order Runge-Kutta method to solve a motion model of an object-parachute system in one embodiment, wherein: Figure 3 (a) is a schematic diagram of the airdrop simulation process in the X direction. Figure 3 (b) is a schematic diagram of the airdrop simulation process in the Y direction. Figure 3 (c) is a schematic diagram of the three-dimensional airdrop trajectory simulation; Figure 4 A schematic top view of an airdrop landing point in one embodiment; Figure 5 Schematic diagram of the implementation steps of a distributed coupling target airdrop method in one embodiment; Figure 6 This is a schematic diagram of the linear model and the drift errors calculated before, during, and around airdrops using this method when the amplitude of the random wind detection error in an experimental simulation is 20%-30% of the detected wind speed; Figure 7 A schematic diagram of the linear model and the drift errors before, during, and around airdrops calculated using this method when the amplitude of the random wind detection error in an experimental simulation is 30%-50% of the detected wind speed; Figure 8 This is a structural block diagram of a distributed coupled target airdrop device based on wind field information in one embodiment; Figure 9 FIG. 1 is a diagram showing the internal structure of a computer device in one embodiment. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical solutions and advantages of this application more clear, the following further describes this application in detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0019] Aiming at the problem that the prediction of airdrop targets in the existing technology is not accurate enough and it is difficult to meet the strict requirements of accurate airdrop in practical applications, this application provides an airdrop target prediction method with dynamic evolution of multi-dimensional parameters driven by multiple models, such as Figure 1 As shown, a method for airdrop target prediction based on dynamic evolution of multi-dimensional parameters driven by multiple models is provided, which includes the following steps: Step S100 : Under the drag model of the cargo-parachute system, a motion model of the cargo-parachute system is constructed by considering the relationship between the parachute deployment area and the drag during the parachute opening process.

[0020] Step S110, obtaining the airdrop height, initial airdrop information and target landing location, and obtaining wind field information of different altitude layers according to the airdrop height.

[0021] Step S120 , based on the cargo-parachute system motion model, the airdrop trajectory is calculated according to the initial airdrop information and the wind field information at different altitudes to obtain the predicted landing location.

[0022] Step S130: Correct the initial airdrop information according to the target landing location and the predicted landing location, and perform the airdrop according to the corrected airdrop information.

[0023] This application presents a distributed coupled airdrop prediction model that comprehensively considers physical phenomena and interactions at different scales to provide an algorithmic model for accurate airdrop motion analysis and guidance strategies. The macroscale includes atmospheric wind direction and speed; the mesoscale includes the shape, size, and material of the payload-parachute system; and the microscale includes the aerodynamic characteristics and drag coefficient of the payload-parachute system affected by the fluid. Inputs at these multiple scales are coupled to influence the model output.

[0024] Furthermore, in this method, after the airdrop height is determined, based on the same wind field information as the linear model, the parachute opening model and drag model of the cargo-parachute system are established to analyze the force of the system in the wind field, and its motion state is analyzed by Newton's second law. Then, the appropriate solution equation is used to obtain the offset landing point of the airdrop and calculate the offset guidance correction amount.

[0025] In step S100, first, the force of the parachute system in the air is analyzed. In this method, the parachute and its load are mainly affected by gravity in the air. , air resistance , and the drag force generated by the relative motion of the wind The drag force Orthogonal decomposition in the X, Y, and Z directions, the force in the X direction is , the force in the Y direction is , the force in the Z direction is , is the gravitational acceleration at the current latitude, and its force analysis is as follows: Figure 2 shown.

[0026] Furthermore, if the air is regarded as a Newtonian fluid, the drag model of the object-parachute system can be expressed as: (1) In formula (1), is the wing drag coefficient, Indicates the air density at the height of the parachute system. represents the equivalent windward diameter of the parachute system, is the background ground wind speed, is the velocity of the cargo-parachute system over the ground.

[0027] Preferably, the wing drag coefficient The reference value can be set to 0.4.

[0028] Specifically, the atmospheric density derived from the Boltzmann energy distribution law is The expression of change with height is as follows: (2) In formula (2), represents the number density of molecules on the Earth's surface, ≈19.52×1024·m -3 , the average radius of the Earth ≈6.371×106m, is the distance from the center of the Earth to the position of the object-parachute system, and the gravitational constant G = 6.67×10 -11 m 3 kg -1 s -2 , Earth's mass ≈5.975×1024kg, Boltzmann constant k=1.3806×10 -23 J / K, T is the thermodynamic temperature, T is taken as the temperature drops by 273.7K for every 100 meters rise, is Avogadro's constant, =6.0220×1023mol -1 , is the molar mass of the gas.

[0029] Furthermore, when analyzing a parachute deployment model, multiple aspects need to be considered, including dynamic analysis of the deployment process, fluid-structure interaction simulation, and optimization of deployment efficiency. In this example, the purpose is to study the impact of wind on the airdrop of a cargo-parachute system, primarily by calculating the motion of the cargo-parachute system in a wind field based on fluid dynamics analysis.

[0030] In conjunction with the force analysis in the drag model of the cargo-parachute system described above, this method focuses on the change in the projected area of the parachute opening during the airdrop process. Knacke's drag area change formula, derived from extensive experimental data, is commonly used in analyzing the force and inflation process of large parachutes. This method uses the Hermite curve to calculate the canopy drag area during the overfill phase. This approximates the change in the projected area of the parachute opening over time as a quadratic curve, expressed as follows: (3) In formula (3), and They respectively represent the moment of releasing the closing and the projection area of the canopy in the stable state. and They correspond to the moment when the canopy is released and the moment when inflation ends respectively.

[0031] Furthermore, the drag model and parachute opening model of the object-parachute system, combined with Newton's second law, can be used to obtain the object-parachute system motion model, that is, the gravity acceleration of the object-parachute system under forces in the X, Y, and Z directions, which can be expressed as: (4) In formula (4), 、 、 They represent the gravitational acceleration of the parachute system under the force in the X, Y, and Z directions respectively. 、 、 are the sum of wind speed and parachute speed in X, Y and Z directions respectively, represents the gravity of the object-parachute system, Indicates the quality of the parachute system.

[0032] Furthermore, the displacements in the three directions can be obtained by integrating the accelerations in the three directions in the time domain.

[0033] In step S110, after obtaining the drop altitude, initial airdrop information, and target landing location for a particular airdrop mission, the aforementioned cargo-parachute system motion model can be used to predict the landing location. The initial airdrop information includes the aircraft's speed at the time of dropping the cargo-parachute system, i.e., the initial velocity of the cargo-parachute system, as well as the time it takes the cargo-parachute system to deploy after being dropped. Because this method takes wind field information into account, and wind field information varies at different altitudes, wind field information for different altitudes is obtained based on the set drop altitude to assist in subsequent accurate airdrop trajectory prediction.

[0034] In step S120, when the predicted landing location is calculated based on the cargo-parachute system motion model, the initial airdrop information, and wind field information at different altitudes, the velocity and displacement at the next moment are calculated based on the cargo-parachute system motion model, the initial airdrop information, and the wind field information at the corresponding altitudes to obtain the position coordinates at the next moment. Based on the current position coordinates, the cargo-parachute system motion model is used to sequentially calculate the position coordinates at the next moments to obtain the airdrop trajectory.

[0035] In this embodiment, when calculating the airdrop trajectory, the position coordinates at a certain moment are expressed as: (5) In formula (5), Indicates the initial position coordinates, or the current position coordinates, Indicates the initial speed, or current speed, Indicates the next moment, represents the acceleration obtained based on the object-parachute system motion model, Represents the integral variable, used to integrate acceleration in the time domain and help construct the relationship between velocity and displacement over time. It is a "virtual" variable introduced to complete the integral mathematically. After the integral is completed, it is eliminated by substituting the upper and lower limits without affecting the final physical result. Its main function is to identify the cumulative calculation of the time dimension during the integral process.

[0036] In this embodiment, when calculating the airdrop trajectory, the Runge-Kutta method is used to calculate the motion process of the distributed coupled parachute system.

[0037] Specifically, assuming the differential equation and initial conditions, it can be expressed as: (6) In formula (6), . Its iterative formula is expressed as: (7) During the airdrop process, the process is very fast, so in order to meet the real-time calculation requirements, the first-order Runge-Kutta method (Euler method) can be used for approximate calculation. The iterative expression is: (8) The above calculation process simulation process, such as Figure 3 shown. Figure 3 , showing the view of a certain airdrop trajectory simulation, in which a height layer is set every 100 meters and the wind field model of each height layer is randomly generated according to the normal distribution. Figure 3 (a) is the side view trajectory in the sailing direction, Figure 3 (b) is the side view trajectory in the vertical navigation direction, Figure 3 (c) is the three-dimensional space trajectory of the airdrop. Figure 4 The relative position distribution of the calculated landing point and the actual landing point in the overlooking direction is shown in the figure.

[0038] Finally, in step S130, after obtaining the predicted airdrop landing location based on the initial airdrop information, the initial airdrop information is corrected based on the target landing location and the predicted landing location. This includes determining the left-right drift distance and the fore-aft drift distance based on the target landing location and the predicted landing location. The initial airdrop velocity in the initial airdrop information and the parachute opening time in the payload-parachute system are then corrected based on the left-right drift distance and the fore-aft drift distance. This correction can be performed directly by obtaining the corresponding correction data based on the difference between the predicted and target values.

[0039] like Figure 5 The figure shows the specific implementation process of the method in this paper.

[0040] This paper also validates the performance of the proposed method through experimental simulation. During the simulation, based on measured meteorological data and field airdrop test results, the performance gain of the optimization method was quantitatively analyzed by comparing the deviations between the predicted and actual results of the two methods. Statistics show that the proposed method improves the accuracy of airdrop impact point slant distance prediction by an average of 52.03%, and the accuracy of the prediction of the left and right distances to the impact point can be improved by an order of magnitude. However, the optimization algorithm also shows that some errors in later batches increase relatively. This is attributed to calculation errors caused by changes in actual wind speed compared to the last wind measurement. Overall, by establishing a distributed coupling model to optimize the airdrop method, its impact point prediction performance can be significantly improved.

[0041] In the experimental simulation, the influence of wind field detection error was analyzed. In order to analyze the stability of the two methods, the relative gain matrix theory (RGA) and Kalman filter algorithm were used to study the influence of wind field detection error on the two method models, analyze the influence mechanism and verify it through numerical simulation.

[0042] Specifically, the Relative Gain Array (RGA) theory is a general method for analyzing the interaction between control variables in a multivariable control system. It helps designers optimize control strategies and improve system stability and performance by quantitatively analyzing the interaction between control variables. The wind field detection error is taken as an input. The RGA theory is used to analyze the impact of the two airdrop model inputs on the landing point output, and then the impact of the wind field detection error on the model performance is analyzed.

[0043] Furthermore, in the linear airdrop prediction model, the input variable is the wind speed detection error at each altitude layer. ; Output is the drift X in the aircraft heading direction k , the vertical heading direction drift is Y k At this time, the transfer function matrix G is: (9) In formula (9): (10) In formula (10), n is the altitude layer, Represents the angle between wind direction and heading. The transfer function matrix is: (11) Then calculate the relative gain matrix RGA using the following formula: (12) Furthermore, the proposed method is subjected to RGA analysis. In this method, the input is the same as the linear model, and the output is the displacement X in the direction parallel to the ground. p Displacement Y from the vertical ground p , transfer function matrix G drag With G sim The expression is the same, this time the horizontal transfer matrix: (13) in, ; In formula (13), and are the components of the wind speed at the i-th altitude layer in the horizontal and vertical directions, and They represent the horizontal and vertical ground direction parachute motion velocity components respectively. In general, they can be At this time, the relative gain matrix of this method is: (14) The diagonal elements of the relative gain matrices of the two models with respect to wind speed detection errors are both 1, indicating that the effects of wind field errors on their respective outputs are independent of each other. At this point, the elements and expressions of the transfer functions of the two models can be examined for analysis.

[0044] The above analysis shows that the transfer matrix of our method considers more complex dynamic principles and has larger element values than the linear model's transfer matrix. Therefore, the distributed coupling method is more sensitive to wind speed detection errors. Furthermore, the two transfer function element expressions show that the sensitivity of the linear model decreases with increasing altitude. However, the distributed coupling method's sensitivity to wind speed detection errors increases with increases in parachute area, air density, and other factors.

[0045] Specifically, Kalman filtering (KF) is a recursive algorithm based on Bayesian estimation, widely used in GNSS and INS data fusion. Its calculation process consists of two main steps: first, a state prediction is made based on the system's output information; second, the measured data is compared with the predicted results and the state estimate is updated using the new measured values.

[0046] The state and measurement model of the Kalman filter is shown below: (15) (16) In formula (15) and formula (16), X k is the state vector, F k is the state transfer matrix; Z k is the measurement vector, Hk is the measurement matrix. k and V k is a zero-mean uncorrelated white noise sequence.

[0047] The Kalman filter is divided into state prediction and state update steps, where the state prediction step can be expressed as: (17) (18) In formula (17) and formula (18), is the prior state estimate vector, is the posterior state estimate at the previous moment, is the prior state covariance matrix, is the posterior state covariance matrix of the previous moment, Q k With R k are the process noise covariance matrix and the measurement noise covariance matrix, respectively.

[0048] The steps for updating the state of the Kalman filter are as follows: (19) (20) (twenty one) When using Kalman filtering for model analysis, the state vector X is specified. k =[x k ,y k , z k , V(x,k), V(y,k), V(z,k)], the state vector includes the aircraft's flight direction displacement x, vertical aircraft's flight direction displacement y, height direction displacement z, aircraft's flight direction wind speed V x , wind speed V in the vertical direction of the aircraft's flight y , height direction wind speed V z .

[0049] It is now planned to use a dual-polarization meteorological radar detection method for airborne ball detection. The characteristic of this method is to enhance the target detection gain by utilizing the correlation characteristics of the airborne sphere in the orthogonal polarization channel. After analyzing the motion equations of the two airdrop models, the corresponding state transfer equation and state transfer matrix can be calculated using the airdrop displacement and wind speed information as the state vector.

[0050] Furthermore, combined with the above analysis of the motion process of the linear model, assuming that the time step is , we can get the state transfer equation of the linear model as: (twenty two) State transition matrix Fk for: (twenty three) For the distributed coupled airdrop prediction model, the drag force of the parachute on the wind should be considered in the model and the dynamic equation should be solved to obtain its state transfer equation. Assuming the time step is , its state transfer equation can be expressed as: (twenty four) Since the Kalman filter requires the state transfer equation to be linear, it is necessary to linearize the nonlinear equation to obtain its Jacobian matrix. After taking partial derivatives of all components in the matrix, the state transfer matrix F is obtained. k As shown below: (25) In order to analyze the impact of wind field detection error on the performance of the two models, let the two model observation matrices H k =I6, where I6 is the 6th-order unit matrix; the same observation noise is introduced during the analysis process By comparing the Kalman gains of the two models under observation noise, the influence of observation error on the two models is analyzed. Since the main focus is on the influence of wind speed observation error and this error is the main factor affecting estimation uncertainty, the process noise Q is not considered when analyzing the Kalman gain. k At the same time, the initial estimation covariance matrix can be made equal to the observation noise covariance matrix, that is, P 0|0 =R k .

[0051] Let the linear model Kalman gain be K kl , in the prediction step: (26) When k=0, since P0|0=Rk, substituting into the above formula we get: (27) In the prediction step of this method: When k=0, due to P 0|0 =R k , after substituting into the above formula, the resulting expression is relatively complex, so in order to simplify the expression, , we can get the formula: (28) Since the two algorithm models are based on the wind speed at the time of airdrop for trajectory prediction, they do not involve the update of wind speed and self-adjustment and observation of parachute status during the airdrop process. That is, there are no new observation values after the initial state in the calculation process of the two models. The error of the initial state can be extended to the error accumulation in the process. Therefore, the state covariance matrix corresponding to the two models in the initial state can be analyzed to determine the impact of observation error on the performance of the two models.

[0052] In the process of analyzing through the state covariance matrix, the diagonal elements P(i,i) represent the state variables x i The estimated variance of the state is larger, and the greater the value, the higher the uncertainty of the state. Therefore, we can compare the diagonal elements P of the two models respectively. 1|0 (1,1), P 1|0 (2,2), P 1|0 The value of (3,3) is used to evaluate the impact of observation error on the displacement prediction performance in three directions; at the same time, the sum of the three elements can also be analyzed to analyze the impact of observation error on the overall displacement prediction performance.

[0053] Taking the analysis of state variable x1 as an example, the two model elements are P 1|0 Dividing the (1,1) term yields the expression From the expression, we can see that since the numerator has the time step term, it is often small when participating in the calculation of the Runge-Kutta type motion equation. The numerator and denominator of the remainder are generally at the same order of magnitude in reality. Therefore, it can be considered that when the time step used in the algorithm calculation is small, the linear model is less sensitive to errors. And from the expression of the diagonal elements in the distributed coupling model and the two models P 1|0 From the (1,1) element division expression, it can be seen that the sensitivity of the distributed coupling model to observation errors is positively correlated with the parachute area and air density. This conclusion is consistent with the analysis through RGA theory.

[0054] Furthermore, in order to verify the conclusions of the above analysis, this paper adds the normal distribution of random detection errors based on the measured wind field data and calculates the airdrop prediction deviation according to the two algorithm models and makes a comparative analysis. The simulation results are as follows: Figure 6-7 shown.

[0055] Figure 6 and Figure 7 The linear model and the drift errors before and after the airdrop and left and right calculated using this method show the amplitude of the random detection error of the wind field when the detected wind speed is 20%-30% and 30%-50%. In the two figures, serial numbers 1-7 are the error results under the wind field data of airdrop day 1, and serial numbers 8-15 are the error results under the wind field data of airdrop day 2.

[0056] In this multi-model-driven airdrop target prediction method with dynamic evolution of multidimensional parameters, a distributed coupled target prediction method for airdrop parachutes in a lidar-detected wind field is proposed by comprehensively analyzing the coupled relationships among wind field distribution, drag, and the parachute deployment process. The Runge-Kutta method is then used to determine the target's final impact point. To validate the performance of this proposed method, field airdrop experiments were conducted using radar-measured meteorological data. The impact of wind field detection errors on the airdrop prediction method was evaluated using relative gain array (RGA) theory and Kalman filtering, and the results were validated using numerical simulations. Experimental results demonstrate that this method achieves higher accuracy than the widely used linear airdrop target prediction method, improving accuracy by 52.03%. Furthermore, the linear prediction method exhibits greater robustness to wind field detection errors.

[0057] It should be understood that although Figure 1 The steps in the flowchart are shown in sequence as indicated by the arrows, but these steps are not necessarily executed in the order indicated by the arrows. Unless otherwise specified in this document, there is no strict order restriction for the execution of these steps, and these steps can be executed in other orders. In addition, Figure 1 At least part of the steps may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily executed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be executed in turn or alternately with other steps or at least part of the sub-steps or stages of other steps.

[0058] In one embodiment, Figure 8 As shown, a distributed coupled target airdrop device based on wind field information is provided, comprising: a cargo-parachute system motion model building module 200, an information acquisition module 210, a landing location prediction module 220, and an airdrop information correction module 230, wherein: The cargo-parachute system motion model building module 200 is used to build the cargo-parachute system motion model under the drag model of the cargo-parachute system, taking into account the relationship between the parachute deployment area and the drag during the parachute opening process; The information acquisition module 210 is used to obtain the airdrop height, initial airdrop information, and target landing location, and obtain wind field information at different altitudes based on the airdrop height; A landing location prediction module 220 is configured to calculate the airdrop trajectory based on the cargo-parachute system motion model, the initial airdrop information, and wind field information at different altitudes to obtain a predicted landing location; The airdrop information correction module 230 is used to correct the initial airdrop information according to the target landing location and the predicted landing location, and perform airdrop according to the corrected airdrop information.

[0059] Regarding the specific definition of the distributed coupled target airdrop device based on wind field information, please refer to the definition of the airdrop target prediction method for the dynamic evolution of multi-dimensional parameters driven by multiple models above, which will not be repeated here. The various modules in the above-mentioned distributed coupled target airdrop device based on wind field information can be implemented in whole or in part by software, hardware, and a combination thereof. The above-mentioned modules can be embedded in or independent of the processor in the computer device in hardware form, or can be stored in the memory of the computer device in software form, so that the processor can call and execute the operations corresponding to the above modules.

[0060] In one embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as follows: Figure 9 As shown. The computer device includes a processor, memory, network interface, display screen and input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and computer program in the non-volatile storage medium. The network interface of the computer device is used to communicate with an external terminal via a network connection. When the computer program is executed by the processor, it implements an airdrop target prediction method with dynamic evolution of multi-dimensional parameters driven by multiple models. The display screen of the computer device can be a liquid crystal display or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a key, trackball or touchpad provided on the computer device housing, or an external keyboard, touchpad or mouse.

[0061] Those skilled in the art will understand that Figure 9 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than shown in the figure, or combine certain components, or have a different component arrangement.

[0062] In one embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented: Under the drag model of the cargo-parachute system, the relationship between the parachute deployment area and the drag during the parachute opening process is considered, and the motion model of the cargo-parachute system is constructed. Obtaining the airdrop height, initial airdrop information, and target landing location, and obtaining wind field information at different altitudes based on the airdrop height; Based on the cargo-parachute system motion model, the airdrop trajectory is calculated according to the initial airdrop information and wind field information at different altitudes to obtain a predicted landing location; The initial airdrop information is corrected according to the target landing location and the predicted landing location, and the airdrop is performed according to the corrected airdrop information.

[0063] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the following steps are implemented: Under the drag model of the cargo-parachute system, the relationship between the parachute deployment area and the drag during the parachute opening process is considered, and the motion model of the cargo-parachute system is constructed. Obtaining the airdrop height, initial airdrop information, and target landing location, and obtaining wind field information at different altitudes based on the airdrop height; Based on the cargo-parachute system motion model, the airdrop trajectory is calculated according to the initial airdrop information and wind field information at different altitudes to obtain a predicted landing location; The initial airdrop information is corrected according to the target landing location and the predicted landing location, and the airdrop is performed according to the corrected airdrop information.

[0064] Those skilled in the art will understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided in this application may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in many forms such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0065] The technical features of the above embodiments can be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0066] The above-described embodiments merely represent several implementation methods of the present application. While the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art could make various modifications and improvements without departing from the spirit of the present application, all of which fall within the scope of protection of the present application. Therefore, the scope of protection of the present patent application shall be determined by the appended claims.

Claims

1. A method for airdrop target prediction based on dynamic evolution of multi-dimensional parameters driven by multiple models, characterized by: The method comprises: Under the drag model of the cargo-parachute system, the relationship between the parachute deployment area and the drag during the parachute opening process is considered, and the motion model of the cargo-parachute system is constructed. Obtaining the airdrop height, initial airdrop information, and target landing location, and obtaining wind field information at different altitudes based on the airdrop height; Based on the cargo-parachute system motion model, the airdrop trajectory is calculated according to the initial airdrop information and wind field information at different altitudes to obtain a predicted landing location; The initial airdrop information is corrected according to the target landing location and the predicted landing location, and the airdrop is performed according to the corrected airdrop information.

2. The airdrop target prediction method based on dynamic evolution of multi-dimensional parameters driven by multiple models according to claim 1 is characterized in that: The drag model of the cargo-parachute system is expressed as: In the above formula, is the wing drag coefficient, Indicates the air density at the height of the parachute system. represents the equivalent windward diameter of the parachute system, is the background ground wind speed, is the velocity of the cargo-parachute system over the ground.

3. The airdrop target prediction method based on dynamic evolution of multi-dimensional parameters driven by multiple models according to claim 2 is characterized in that: The object-parachute system motion model is expressed as: In the above formula, 、 、 They represent the gravitational acceleration of the parachute system under the force in the X, Y, and Z directions respectively. 、 、 are the sum of wind speed and parachute speed in X, Y and Z directions respectively, represents the gravity of the object-parachute system, Indicates the quality of the parachute system.

4. The airdrop target prediction method based on dynamic evolution of multi-dimensional parameters driven by multiple models according to any one of claims 1 to 3, characterized in that: When the airdrop trajectory is calculated based on the cargo-parachute system motion model, the initial airdrop information, and the wind field information at different altitudes to obtain the predicted landing location: Based on the cargo-parachute system motion model, the velocity and displacement at the next moment are calculated according to the initial airdrop information and the wind field information of the corresponding altitude layer to obtain the position coordinates at the next moment; According to the position coordinates at the current moment, the position coordinates at the next moment are calculated in sequence using the object-parachute system motion model to obtain the airdrop trajectory.

5. The airdrop target prediction method based on dynamic evolution of multi-dimensional parameters driven by multiple models according to claim 4 is characterized in that: When calculating the airdrop trajectory, the position coordinates at a certain moment are expressed as: In the above formula, Indicates the initial position coordinates, or the current position coordinates, Indicates the initial speed, or current speed, Indicates the next moment, represents the acceleration obtained based on the object-parachute system motion model, Represents the integral variable, which is used to integrate the acceleration in the time domain.

6. The airdrop target prediction method based on dynamic evolution of multi-dimensional parameters driven by multiple models according to claim 5 is characterized in that: When calculating the airdrop trajectory, the Runge-Kutta method is used.

7. The airdrop target prediction method based on dynamic evolution of multi-dimensional parameters driven by multiple models according to claim 6 is characterized in that: The correcting the airdrop initial information according to the target landing location and the predicted landing location includes: According to the target landing location and the predicted landing location, a left-right drift distance and a front-back drift distance are obtained; According to the left-right drift distance and the front-back drift distance, the initial airdrop speed of the initial airdrop information and the parachute opening time in the cargo-parachute system are corrected.

8. An airdrop target prediction device with dynamic evolution of multi-dimensional parameters driven by multiple models, characterized by: The device comprises: The cargo-parachute system motion model construction module is used to consider the relationship between the parachute deployment area and the drag during the parachute opening process under the drag model of the cargo-parachute system, and to construct the cargo-parachute system motion model; An information acquisition module is used to obtain the airdrop height, initial airdrop information, and target landing location, and obtain wind field information at different altitudes based on the airdrop height; A landing location prediction module is used to calculate the airdrop trajectory based on the cargo-parachute system motion model, the initial airdrop information, and wind field information at different altitudes to obtain a predicted landing location; The airdrop information correction module is used to correct the initial airdrop information according to the target landing location and the predicted landing location, and perform airdrop according to the corrected airdrop information.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 7 are implemented.

Images