Multi-modal low altitude flight fusion method based on data analysis
By performing sliding slicing and spatiotemporal curl vector analysis on the low-altitude flight trajectory dataset, an evolution equation for the low-altitude flight trajectory was established, and a coupled topology graph was generated. This solved the real-time and visualization problems of dynamic prediction of low-altitude flight trajectories, and improved the prediction accuracy and efficiency in complex airspace.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- GUANGDONG LOW-ALTITUDE ECONOMIC IND DEVELOPMENT CO LTD
- Filing Date
- 2026-03-04
- Publication Date
- 2026-06-05
AI Technical Summary
Existing technologies focus on static analysis of low-altitude flight trajectories, lacking dynamic analysis of flight airspace. This makes it difficult to predict the evolution of low-altitude flight trajectories in real time, especially in complex airspace conditions where processing efficiency is low and it is difficult to provide accurate and dynamic flight information visualization data.
By collecting low-altitude flight trajectory datasets, we generate spatiotemporal slices through sliding slicing, calculate the spatiotemporal curl vector and flight airspace distortion, establish the evolution equation of low-altitude flight trajectories, and combine the spatiotemporal curl vector constraints to generate a coupled topology graph for data fusion.
It enables dynamic prediction and visualization of low-altitude flight trajectories, improving real-time performance and accuracy in complex airspace, and solving the limitations of static analysis and insufficient prediction accuracy of traditional methods.
Smart Images
Figure CN122154456A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of data processing technology, and in particular to a multimodal low-altitude flight fusion method based on data analysis. Background Technology
[0002] Multimodal low-altitude flight information refers to data and information related to the field of low-altitude flight, including flight plans, low-altitude flight routes, weather conditions, aircraft performance, flight delays, and flight safety records.
[0003] Low-altitude flight data fusion presents complex flight data graphically, making it more intuitive and understandable. The aviation field involves a large amount of dynamic and complex information, and traditional text or numerical presentation methods are insufficient to effectively convey key information. Through visualization, decision-makers can quickly understand flight operation status, weather changes, and other factors, enabling timely adjustments and responses, thus improving the safety and efficiency of air transport. Furthermore, low-altitude flight data fusion can also help passengers understand flight dynamics in real time.
[0004] However, existing technologies typically focus on static analysis of low-altitude flight trajectories, lacking dynamic analysis of flight airspace and making it difficult to predict the evolution of low-altitude flight trajectories in real time. Furthermore, existing methods are inefficient when dealing with complex airspace conditions, especially in areas with dense flight, making it difficult to provide accurate and dynamic flight information visualization data. Summary of the Invention
[0005] In view of the shortcomings of the prior art, the purpose of this invention is to provide a multimodal low-altitude flight fusion method based on data analysis. This method can solve the problem that existing technologies typically focus on static analysis of low-altitude flight trajectories, lack dynamic analysis of flight airspace, and are difficult to predict the evolution of low-altitude flight trajectories in real time. Furthermore, existing methods are inefficient when dealing with complex airspace conditions, especially in densely populated airspaces, making it difficult to provide accurate and dynamic flight information visualization data.
[0006] This invention proposes a multimodal low-altitude flight fusion method based on data analysis, including: S1: Collect a dataset of low-altitude flight trajectories in the target airspace; S2: Perform sliding slicing on the low-altitude flight trajectory dataset to generate multiple spatiotemporal slices; S3: For each spatiotemporal slice, calculate the spatiotemporal curl vector describing the degree of distortion of the low-altitude flight trajectory by combining the low-altitude flight trajectory dataset, and determine the degree of distortion of the flight airspace based on the spatiotemporal curl vector; S4: Combine the flight airspace distortion to verify whether the spacetime curl vector gradient satisfies the boundedness. If yes, proceed to step S5; otherwise, reduce the mesh size of the spacetime slice and return to step S3. S5: Establish the low-altitude flight trajectory evolution equation with the spatiotemporal curl vector as a constraint; S6: Outputs predicted flight trajectory based on the low-altitude flight trajectory evolution equation; S7: Overlay the flight airspace distortion and flight prediction trajectory to generate a coupled topology map and complete the fusion of low-altitude flight data.
[0007] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following: In this embodiment of the invention, the dynamic distortion characteristics of the airspace are quantized by spatiotemporal curl vector quantization, and multi-scale spatiotemporal data analysis is achieved by combining sliding slicing technology. Boundedness verification drives the adaptive grid optimization algorithm, effectively balancing computational accuracy and efficiency. The trajectory evolution equation based on physical constraints constructs the coupling relationship between airspace state and trajectory prediction. The visualization fusion of dynamic airspace characteristics and predicted trajectory is achieved through the coupled topology graph. This solves the problems of limitations of static analysis, insufficient prediction accuracy, and low efficiency in processing complex airspace in traditional methods, significantly improving the real-time performance of low-altitude flight trajectory evolution prediction, the accuracy of airspace situational awareness, and the intuitive visualization effect. Attached Figure Description
[0008] The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. Throughout the drawings, the same reference numerals denote the same parts. Obviously, the drawings described below are merely some embodiments of the present invention, and those skilled in the art can obtain other drawings based on these drawings without any creative effort.
[0009] Figure 1 This is a flowchart illustrating the multimodal low-altitude flight fusion method based on data analysis provided in an embodiment of the present invention. Detailed Implementation
[0010] To enable those skilled in the art to better understand the technical solutions in the embodiments of the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. It should be understood that these descriptions are merely exemplary and are not intended to limit the scope of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.
[0011] The multimodal low-altitude flight fusion method based on data analysis provided by the present invention will be described in detail below with reference to the accompanying drawings, through specific embodiments and application scenarios.
[0012] Reference manual attached Figure 1The diagram illustrates a flowchart of the multimodal low-altitude flight fusion method based on data analysis provided in an embodiment of the present invention.
[0013] This invention provides a multimodal low-altitude flight fusion method based on data analysis, including: S1: Collect a dataset of low-altitude flight trajectories in the target airspace.
[0014] The target airspace refers to the area or air corridor where low-altitude flights occur within a specific time period. These areas involve multiple low-altitude flight paths. Low-altitude flight trajectory datasets contain data such as position, speed, time, and heading during low-altitude flight. This data can be collected in real time via satellite or radar systems, forming continuous records of low-altitude flight trajectories. By collecting low-altitude flight trajectory datasets from the target airspace, multiple low-altitude flight information can be obtained, providing data support for subsequent analysis and prediction, ensuring accurate monitoring and analysis of flight dynamics in both spatiotemporal dimensions.
[0015] S2: Perform sliding slicing on the low-altitude flight trajectory dataset to generate multiple spatiotemporal slices.
[0016] Spatiotemporal slicing refers to dividing the low-altitude flight trajectory dataset into two dimensions: time and space. Using a sliding window method, the entire low-altitude flight trajectory data is divided into multiple smaller time segments, and the target airspace is also divided into several regions. Each spatiotemporal slice represents the state of the flight at a specific moment and within its corresponding region. By segmenting the low-altitude flight trajectory data by time and space using the sliding slicing method, the dynamic changes of the flight within a specific time period can be better captured, providing detailed local data for subsequent trajectory analysis, distortion calculation, and other tasks.
[0017] In one possible implementation, S2 specifically refers to: Spatiotemporal slices are obtained by sliding slices of the low-altitude flight trajectory dataset according to a fixed time window.
[0018] The size of the fixed time window is the ratio of the acquisition duration corresponding to the low-altitude flight trajectory dataset to the preset number of spatiotemporal slices. Those skilled in the art can set the size of the preset number of spatiotemporal slices according to actual needs; this invention does not impose any limitations on this.
[0019] It should be noted that by using the sliding slicing technique with a fixed time window, continuous low-altitude flight trajectory data is dynamically divided into continuous overlapping spatiotemporal slices, enabling time-segmented analysis of dynamic changes in the airspace. This ensures both data timeliness and the preservation of trajectory continuity through window overlap, providing high-resolution spatiotemporal data for subsequent curl calculations and trajectory predictions.
[0020] S3: For each spatiotemporal slice, calculate the spatiotemporal curl vector describing the degree of distortion of the low-altitude flight trajectory by combining the low-altitude flight trajectory dataset, and determine the degree of distortion of the flight airspace based on the spatiotemporal curl vector.
[0021] The spatiotemporal curl vector is a vector field constructed by quantifying the rotational characteristics (such as nonlinear deformations like bending and twisting) of low-altitude flight trajectories in the spatiotemporal field, reflecting the local intensity and direction of dynamic distortion of the trajectory. The flight airspace distortion is a global statistical representation of the spatiotemporal curl vector. By integrating the curl information of multiple trajectories, it quantifies the overall dynamic distortion of the airspace (such as congested areas and abrupt changes in the flow field), providing key physical constraints for trajectory evolution prediction and visualization.
[0022] By capturing the dynamic distortion features of the trajectory through spatiotemporal curl vectors and combining them with spatial distortion quantification to quantify the overall distortion state, physical constraints are provided for trajectory prediction and visualization, enabling accurate analysis and efficient identification of complex spatial situations.
[0023] In one possible implementation, S3 specifically includes: S301: Grid the spatiotemporal slice.
[0024] S302: Calculate the average flight speed of each aircraft within each grid.
[0025] The formula for calculating average flight speed is as follows: .
[0026] in, and They represent flight k In time interval within x Directional displacement and y Directional displacement, Indicates falling into the grid ( i , j The flight group, Represents a grid ( i , j ) within Average flight speed at any given time Represents a grid ( i , j The number of flights within ).
[0027] in, The time point refers to the discrete sampling time within the fixed time window to which the spatiotemporal slice belongs, and the time interval is... For calculation The time span of the instantaneous velocity, which is less than or equal to the time interval between adjacent discrete sampling moments.
[0028] S303: Based on the average flight speed, calculate the spatiotemporal curl vector and the flight airspace distortion degree, which describe the degree of distortion of the low-altitude flight trajectory. The flight airspace distortion degree is the z-direction component of the spatiotemporal curl vector.
[0029] The z-direction component is the component of the spacetime curl vector in the direction perpendicular to the xy plane, which characterizes the rotational intensity of the velocity field in the space (such as dynamic distortions like vortices and shear flow).
[0030] In flight scenarios, the spatiotemporal curl vector can capture hidden airspace conflicts that traditional methods (such as trajectory density) cannot reflect (such as "vortices" formed by flight circling and waiting, and "shear flows" caused by multi-directional intersections).
[0031] Specifically, this process discretizes the airspace through gridding, calculates the average velocity vector field of flight within each grid using a sliding time window, and then quantifies the rotational characteristics of low-altitude flight trajectories in the airspace (such as vortices, shear flow, and other dynamic distortions) through curl calculation. Finally, the z-direction curl component is extracted as an index of airspace distortion. Gridding and a sliding time window enable spatiotemporal multi-scale analysis, taking into account both local dynamics and overall trends. Curl calculation based on the average velocity field can capture implicit conflicts that traditional trajectory density methods cannot identify (such as vortices formed by circling and shear flow caused by multi-directional intersections), improving the detection accuracy of airspace anomalies. Using the z-direction curl component to characterize airspace distortion directly correlates with trajectory deformation intensity, providing physical constraints for trajectory prediction and significantly enhancing the predictive reliability and visualization-based decision support capabilities for low-altitude flight trajectory evolution under complex airspace conditions.
[0032] In one possible implementation, S303 specifically includes: S3031: Determine the average flight speed in the x-direction and the average flight speed in the y-direction within the corresponding grid based on the average flight speed.
[0033] Among them, the average flight velocity in the x-direction and the average flight velocity in the y-direction are the velocity components of flight in the horizontal plane within the grid, reflecting the motion trend of flight in two-dimensional space.
[0034] S3032: Based on the average flight velocity in the x-direction and the average flight velocity in the y-direction, establish a virtual velocity field that describes the motion trend of the aircraft in the horizontal plane within the grid.
[0035] The formula for the virtual velocity field is as follows: .
[0036] in, These represent the average flight speed in the x-direction and the average flight speed in the y-direction, respectively. This represents a virtual velocity field.
[0037] S3033: Calculate the virtual velocity field and apply the curl operator to obtain the spatiotemporal curl vector describing the degree of distortion of the low-altitude flight trajectory.
[0038] The formula for the spacetime curl vector is as follows: .
[0039] in, express Time grid ( i , j The spacetime curl vector of ) Represents the vector differential operator. Represents the curl operator, This indicates the partial derivative. These represent the unit vectors in the x-direction, y-direction, and time direction, respectively. Let represent the partial derivatives with respect to the x-direction, the partial derivatives with respect to the y-direction, and the partial derivatives with respect to the time variable, respectively. t The partial derivatives of .
[0040] S3034: Calculate the z-direction component of the spacetime curl vector to obtain the flight airspace distortion.
[0041] The specific formula for calculating the distortion of the flight airspace is as follows: .
[0042] in, The z-direction component of the spacetime curl vector represents the distortion of the airspace during flight.
[0043] Specifically, this process constructs a virtual velocity field containing a time dimension by statistically averaging the velocity components in a gridded manner, and uses a curl operator to quantify the rotational characteristics of the velocity field in the airspace. Its z-direction component directly reflects the intensity of dynamic distortions in the airspace (such as vortices and shear flows). By introducing a virtual velocity field with a time dimension, the limitations of traditional two-dimensional velocity fields are overcome, enabling spatiotemporal coupled dynamic distortion modeling. The curl operator can accurately capture the nonlinear deformation characteristics of low-altitude flight trajectories (such as circling and waiting, and intersection conflicts), compensating for the deficiencies of static indicators such as trajectory density. The z-direction component, as an indicator of airspace distortion, is directly related to physical constraints, providing a high-precision dynamic airspace state representation for trajectory prediction, significantly improving the visualization analysis and decision support capabilities under complex airspace situations.
[0044] S4: Combine the flight airspace distortion to verify whether the spacetime curl vector gradient satisfies boundedness. If yes, proceed to step S5; otherwise, reduce the mesh size of the spacetime slice and return to step S3.
[0045] The spacetime curl vector gradient is a tensor field describing the rate of change of the spacetime curl vector in the spatial and temporal dimensions. It reflects the local dynamic evolution rate of spatial distortion characteristics. An excessively large gradient magnitude may lead to numerical instability or distortion of physical meaning. The verification module dynamically adjusts the mesh size of the spacetime slice by determining whether the curl gradient satisfies boundedness, ensuring that the rate of change of the curl field is within a reasonable range. This improves computational stability and the accuracy of spatial distortion modeling, providing a reliable physical constraint basis for the trajectory evolution equation.
[0046] In one possible implementation, S4 specifically includes: S401: Calculate the spacetime curl vector gradient by combining the flight airspace distortion.
[0047] The formula for calculating the spacetime curl vector gradient is as follows: .
[0048] in, This represents the gradient of the spacetime curl vector.
[0049] S402: Calculate the Frobenius norm of the spacetime curl vector gradient.
[0050] S403: Calculate the instantaneous radius of curvature of the trajectory of each aircraft within a spacetime slice.
[0051] .
[0052] in, Indicates the first k A flight The instantaneous radius of curvature of the trajectory at time t. and They represent flight k of x Direction coordinates and y Direction coordinates.
[0053] S404: The average radius of curvature of the trajectory is obtained by summing the instantaneous trajectory curvature radii according to the number of aircraft and taking the average.
[0054] .
[0055] in, This represents the average radius of curvature of the trajectory. N This represents the total number of flights within a spacetime slice.
[0056] S405: Calculate the standard deviation of the instantaneous trajectory radius of curvature.
[0057] .
[0058] in, It represents the standard deviation.
[0059] S406: Combining the Frobenius norm, mean radius of curvature of the trajectory, and standard deviation of the spacetime curl vector gradient, determine whether the spacetime curl vector gradient satisfies boundedness. If the Frobenius norm is less than or equal to the discriminant value, the output spacetime curl vector gradient satisfies boundedness; otherwise, the output spacetime curl vector gradient does not satisfy boundedness. The discriminant value is specifically the product of a first ratio and a second ratio. The first ratio is specifically the ratio between the standard deviation and the mean radius of curvature of the trajectory, and the second ratio is specifically the second ratio between the aircraft density and the preset safety interval.
[0060] The specific formula for boundedness verification is as follows: .
[0061] in, The Frobenius norm represents the gradient of the spacetime curl vector. This represents the density of aircraft within the corresponding time slice. This represents the minimum allowable time difference between adjacent flights passing through the same grid.
[0062] Specifically, this verification process comprehensively judges the boundedness of the spatiotemporal curl vector gradient through multi-dimensional indicators: First, the Frobenius norm of the gradient matrix (reflecting the overall strength of the gradient) is calculated. This is combined with the statistical characteristics of the instantaneous radius of curvature of the trajectory (mean and standard deviation characterize the concentration and dispersion of trajectory curvature), and the ratio of aircraft density to safety clearance (reflecting the correlation between airspace congestion and safety threshold) is introduced. Finally, a discriminant formula couples physical constraints (curvature characteristics) with airspace state (density / safety clearance) to dynamically assess whether the gradient is within a reasonable range. The stability of trajectory deformation is quantified through curvature statistics (mean, standard deviation), avoiding the one-sidedness of a single indicator. The introduction of the ratio of aircraft density to safety clearance establishes a physical correlation between airspace congestion and gradient constraints, improving the dynamic adaptability of the discriminant threshold. The joint judgment mechanism of the Frobenius norm and multi-dimensional indicators effectively filters abnormal gradient fluctuations, ensuring that the rate of change of the curl field is within a safe range, providing a stable and reliable constraint basis for the trajectory evolution equation, thereby improving the robustness and physical rationality of trajectory prediction in complex airspace.
[0063] S5: Establish the evolution equation of low-altitude flight trajectory using the spatiotemporal curl vector as a constraint.
[0064] The low-altitude flight trajectory evolution equation is a dynamic differential equation constructed based on the spatiotemporal curl vector. By using airspace distortion characteristics (such as nonlinear deformations like rotation and stretching) as constraints, and combining the historical motion state (velocity, direction, acceleration) of the low-altitude flight trajectory with real-time airspace flow field parameters (wind speed, congestion, control rules), a predictive model of trajectory changes over time and space is established. By introducing physical constraints, the trajectory evolution process of flight in complex airspace can be dynamically simulated, thereby improving the physical rationality and airspace adaptability of the prediction results. By embedding the spatiotemporal curl vector as a physical constraint into the trajectory evolution model, the predicted trajectory strictly follows the dynamic distortion laws of the airspace, effectively avoiding trajectory deviations caused by neglecting changes in the airspace flow field in traditional methods, and significantly improving the accuracy and stability of low-altitude flight trajectory prediction in dense airspace.
[0065] In one possible implementation, the low-altitude flight trajectory evolution equation is specifically a second-order dynamic model of the low-altitude flight trajectory under the combined influence of the spatiotemporal curl vector and the control instructions provided by the air traffic control system.
[0066] The formula for the second-order dynamic model is as follows: .
[0067] in, Indicates flight k trajectory For time variables t The second derivative, This represents a control command vector that includes heading angle adjustments, speed adjustments, and altitude adjustments. and These represent the curl influence coefficient and the control weight coefficient, respectively.
[0068] Optionally, and Both can be set to 0.5. Trajectory The complete representation is .
[0069] Specifically, this process constructs a second-order dynamic model, using the spatiotemporal curl vector (representing dynamic distortion of the airspace) and air traffic control commands (heading / speed / altitude adjustments) as dual driving factors to simulate the acceleration changes of low-altitude flight trajectories. The curl term reflects the nonlinear disturbance of airspace distortion on the trajectory, while the control term embodies the rigid constraints of human intervention. Weight adjustment makes the model both physically adaptable and rule-compliant. The second-order derivative model captures the dynamic acceleration of the trajectory, outperforming traditional first-order models. The coupling mechanism between the curl and control terms adapts to abrupt changes in the airspace flow field (such as circling and shear flow) while strictly adhering to air traffic control rules, significantly improving the physical rationality of trajectory prediction and the feasibility of real-time scheduling in complex airspaces.
[0070] S6: Outputs the predicted flight trajectory based on the low-altitude flight trajectory evolution equation.
[0071] The flight prediction trajectory is a dynamic flight path simulated over a future time period based on evolutionary equations. It comprehensively considers airspace distortion characteristics under spatiotemporal curl vector constraints, historical trajectory parameters, and real-time flow field changes, generating physically plausible trajectory prediction results through numerical solutions. By dynamically linking airspace distortion and trajectory evolution laws through evolutionary equations, the predicted trajectory strictly adapts to dynamic airspace constraints, significantly improving the accuracy and stability of low-altitude flight trajectory prediction in complex airspaces, and providing a reliable basis for real-time scheduling and visual decision-making.
[0072] In one possible implementation, S6 specifically includes: S601: Obtain the initial flight state, prediction duration, and prediction time step. The initial flight state includes the current time, current position, current speed, and current air traffic control command vector.
[0073] Optionally, the prediction duration can be set to 15 minutes, and the prediction time step can be limited to 10 seconds, meaning that a predicted location point is output every 10 seconds.
[0074] S602: Input the initial flight state into the low-altitude flight trajectory evolution equation, solve the low-altitude flight trajectory evolution equation, and obtain the predicted flight trajectory.
[0075] The solution process involves calculating the future position step-by-step using the RK4 method to generate the predicted flight trajectory. The specific process is as follows: For the second-order dynamic equation: (in, (), which can be broken down into a system of first-order equations: .
[0076] For each time step: ,in, The prediction time step is calculated through the following four iterative steps. and : .
[0077] .
[0078] .
[0079] .
[0080] Last Update: .
[0081] .
[0082] in, n The time step markers are represented as integers, and all unknowns involved are intermediate variables.
[0083] The result Updated to Repeat the iteration, recording all time steps. This yields the predicted flight trajectory.
[0084] Specifically, this process achieves high-precision prediction of low-altitude flight trajectories by decomposing the second-order dynamic equations into a system of first-order equations and iteratively solving them step-by-step using the fourth-order Runge-Kutta method (RK4). The RK4 method significantly reduces numerical errors and improves the accuracy and stability of trajectory prediction through a weighted average of four intermediate calculations, making it particularly suitable for nonlinear dynamic systems coupled with airspace curl terms and air traffic control instructions. By treating curl effects and air traffic control instructions as dual driving factors, the predicted trajectory adheres to both airspace dynamic distortion laws (such as vortices and shear flows) and strictly satisfies air traffic control constraints, significantly enhancing the physical rationality and scheduling feasibility of trajectory prediction in complex airspaces, and providing a reliable basis for low-altitude flight data fusion and real-time decision-making.
[0085] S7: Overlay the flight airspace distortion and flight prediction trajectory to generate a coupled topology map and complete the fusion of low-altitude flight data.
[0086] The coupled topology graph is a visual representation that integrates the spatial distribution characteristics of airspace distortion with the dynamic evolution path of the predicted trajectory in the spatiotemporal dimension. It links airspace distortion regions with trajectory flow directions through topological structure, forming a synergistic mapping relationship between airspace state and flight behavior. By coupling airspace distortion with the predicted trajectory, the dynamic airspace characteristics and future trajectories are visualized synchronously, significantly improving the real-time perception and scheduling decision-making capabilities for complex airspace situations.
[0087] In one possible implementation, S7 specifically includes: S701: Determine the flight airspace distortion threshold, wherein the flight airspace distortion threshold is a preset proportional quantile of the flight airspace distortion.
[0088] It should be noted that those skilled in the art can set the size of the preset proportional quantile according to actual needs, and this invention does not limit this.
[0089] S702: Extract the flight airspace distortion isosurface from each spatiotemporal slice, which is consistent with the flight airspace distortion threshold.
[0090] S703: Project the predicted flight trajectory onto the flight airspace distortion isosurface to obtain a coupled topology map.
[0091] S704: Color mapping is performed on the coupled topology map to complete the fusion of low-altitude flight data.
[0092] Specifically, this process dynamically sets the quantile threshold for airspace distortion, extracts corresponding isosurfaces (such as high-distortion regions), projects the predicted trajectory onto the isosurface to generate a topology map, and visually presents the relationship between airspace status and trajectory through color mapping. The quantile threshold adapts to different airspace density scenarios, avoiding misjudgments caused by fixed thresholds. Isosurface extraction can accurately identify areas with high airspace distortion (such as vortices and shear flows), assisting in identifying potential conflict risks. Trajectory-airspace coupling mapping: By projecting, the trajectory is associated with the airspace distortion feature space, visually revealing the physical mechanism by which the trajectory is affected by airspace dynamics. Color mapping simultaneously presents the distortion intensity and trajectory distribution, improving the perception efficiency and decision support capability of complex airspace situations, significantly outperforming traditional static trajectory overlay methods.
[0093] In one possible implementation, S704 specifically includes: S7041: Calculate the flight airspace distortion modulus in the coupled topology graph.
[0094] S7042: Color mapping is performed based on the magnitude of the flight airspace distortion modulus using color numerical mapping rules.
[0095] The color-coded numerical mapping rule maps the magnitude of airspace distortion to a color spectrum (e.g., blue-green-red) based on intensity gradients. Warm and cool hues visually distinguish the degree of airspace distortion, highlighting high-distortion areas in red and low-distortion areas in blue, thus visually linking airspace status with trajectory distribution. Quantifying the intensity of dynamic airspace distortion through color-coded mapping makes the coupling relationship between complex airspace features and predicted trajectories readily apparent, significantly improving situational awareness efficiency and the intuitiveness of scheduling decisions.
[0096] In one possible implementation, it also includes: The low-altitude flight trajectory dataset is updated at preset intervals.
[0097] Understandably, by regularly refreshing the low-altitude flight trajectory dataset, the system can ensure that it is always based on the latest dynamic airspace data for analysis, thereby improving the real-time performance and accuracy of trajectory prediction and airspace visualization results, and enhancing the system's dynamic response capability to complex airspace changes.
[0098] In practical applications, the system's acquisition module obtains a dataset of low-altitude flight trajectories in the target airspace. The slicing module then divides the data into spatiotemporal slices through sliding slicing for detailed analysis of the low-altitude flight trajectories. The computation module uses spatiotemporal curl vectors to calculate the airspace distortion, capturing the dynamic distortion characteristics of low-altitude flight trajectories and further predicting their evolution by combining physical constraints. The verification module ensures that the curl vector gradient satisfies boundedness, guaranteeing computational stability. Finally, the output module generates high-precision flight prediction trajectories, and the overlay module combines the flight prediction trajectories with airspace distortion to generate a coupled topology graph for visualization. The system's advantage lies in its precise modeling of spatiotemporal curl vectors and airspace distortion, enabling better revelation of potential conflicts and dynamic changes in complex airspaces. This improves the accuracy and real-time performance of low-altitude flight trajectory prediction and scheduling decisions, significantly enhancing the visualization and decision support capabilities of flight information, and adapting to the needs of dense airspaces and complex low-altitude flight environments.
[0099] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following: In this embodiment of the invention, the dynamic distortion characteristics of the airspace are quantized by spatiotemporal curl vector quantization, and multi-scale spatiotemporal data analysis is achieved by combining sliding slicing technology. Boundedness verification drives an adaptive grid optimization algorithm, effectively balancing computational accuracy and efficiency. A trajectory evolution equation based on physical constraints constructs the coupling relationship between airspace state and trajectory prediction. A coupled topology graph is used to achieve the visual fusion of dynamic airspace features and predicted trajectories. This solves the problems of limitations in static analysis, insufficient prediction accuracy, and low efficiency in handling complex airspace in traditional methods, significantly improving the real-time performance of low-altitude flight trajectory evolution prediction, the accuracy of airspace situational awareness, and the intuitive visualization effect.
[0100] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the embodiments of the present invention, and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the protection scope of the present invention.
Claims
1. A multimodal low-altitude flight fusion method based on data analysis, characterized in that, include: S1: Collect a dataset of low-altitude flight trajectories in the target airspace; S2: Perform sliding slicing on the low-altitude flight trajectory dataset to generate multiple spatiotemporal slices; S3: For each of the aforementioned spatiotemporal slices, calculate the spatiotemporal curl vector describing the degree of distortion of the low-altitude flight trajectory by combining the low-altitude flight trajectory dataset, and determine the degree of distortion of the flight airspace based on the spatiotemporal curl vector; S4: Combine the flight airspace distortion to verify whether the spacetime curl vector gradient satisfies boundedness. If yes, proceed to step S5; otherwise, reduce the grid size of the spacetime slice and return to step S3. S5: Using the spatiotemporal curl vector as a constraint, establish the evolution equation of the low-altitude flight trajectory; S6: Output the predicted flight trajectory based on the low-altitude flight trajectory evolution equation; S7: Superimpose the flight airspace distortion and the flight prediction trajectory to generate a coupled topology map and complete the low-altitude flight data fusion.
2. The multimodal low-altitude flight fusion method based on data analysis according to claim 1, characterized in that, Specifically, S2 is: The spatiotemporal slice is obtained by sliding slices of the low-altitude flight trajectory dataset according to a fixed time window.
3. The multimodal low-altitude flight fusion method based on data analysis according to claim 1, characterized in that, S3 specifically includes: S301: Divide each of the spatiotemporal slices into a grid; S302: Calculate the average flight speed of each aircraft within each grid; S303: Based on the average flight speed, calculate the spatiotemporal curl vector describing the degree of distortion of the low-altitude flight trajectory and the flight airspace distortion, wherein the flight airspace distortion is the z-direction component of the spatiotemporal curl vector.
4. The multimodal low-altitude flight fusion method based on data analysis according to claim 3, characterized in that, Specifically, S303 includes: S3031: Determine the average flight speed in the x-direction and the average flight speed in the y-direction within the corresponding grid based on the average flight speed; S3032: Based on the average flight velocity in the x-direction and the average flight velocity in the y-direction, establish a virtual velocity field describing the motion trend of the aircraft in the horizontal plane within the grid; S3033: Calculate the virtual velocity field and apply the curl operator to obtain the spatiotemporal curl vector describing the degree of distortion of the low-altitude flight trajectory; S3034: Calculate the z-direction component of the spatiotemporal curl vector to obtain the flight airspace distortion.
5. The multimodal low-altitude flight fusion method based on data analysis according to claim 1, characterized in that, S4 specifically includes: S401: Calculate the spacetime curl vector gradient based on the aforementioned airspace distortion. S402: Calculate the Frobenius norm of the spacetime curl vector gradient; S403: Calculate the instantaneous radius of curvature of the trajectory of each aircraft within the spatiotemporal slice; S404: Sum the instantaneous trajectory curvature radii according to the number of aircraft and take the average to obtain the average trajectory curvature radius; S405: Calculate the standard deviation of the instantaneous trajectory radius of curvature; S406: Combining the Frobenius norm of the spacetime curl vector gradient, the mean radius of curvature of the trajectory, and the standard deviation, determine whether the spacetime curl vector gradient satisfies boundedness. If the Frobenius norm is less than or equal to the discriminant value, output that the spacetime curl vector gradient satisfies boundedness; otherwise, output that the spacetime curl vector gradient does not satisfy boundedness. The discriminant value is specifically the product of a first ratio and a second ratio. The first ratio is specifically the ratio between the standard deviation and the mean radius of curvature of the trajectory, and the second ratio is specifically the second ratio between the aircraft density and the preset safety interval.
6. The multimodal low-altitude flight fusion method based on data analysis according to claim 1, characterized in that, The evolution equation of the low-altitude flight trajectory is specifically a second-order dynamic model of the low-altitude flight trajectory under the combined influence of the spatiotemporal curl vector and the control commands provided by the air traffic control system.
7. The multimodal low-altitude flight fusion method based on data analysis according to claim 1, characterized in that, S6 specifically includes: S601: Obtain the initial flight state, prediction duration, and prediction time step, wherein the initial flight state includes the current time, current position, current speed, and current air traffic control command vector; S602: Input the initial flight state into the low-altitude flight trajectory evolution equation, and solve the low-altitude flight trajectory evolution equation to obtain the predicted flight trajectory.
8. The multimodal low-altitude flight fusion method based on data analysis according to claim 1, characterized in that, Specifically, S7 includes: S701: Determine the flight airspace distortion threshold, wherein the flight airspace distortion threshold is a preset proportional quantile of the flight airspace distortion; S702: Extract flight airspace distortion isosurfaces that are consistent with the flight airspace distortion threshold from each of the spatiotemporal slices; S703: Project the predicted flight trajectory onto the flight airspace distortion isosurface to obtain the coupled topology map; S704: Perform color mapping on the coupled topology map to complete the fusion of low-altitude flight data.
9. The multimodal low-altitude flight fusion method based on data analysis according to claim 8, characterized in that, Specifically, S704 includes: S7041: Calculate the flight airspace distortion modulus value in the coupled topology diagram; S7042: Color mapping is performed based on the magnitude of the flight airspace distortion modulus using color numerical mapping rules.
10. The multimodal low-altitude flight fusion method based on data analysis according to claim 1, characterized in that, Also includes: The low-altitude flight trajectory dataset is updated at preset intervals.