Method, device and equipment for dynamically planning real-time route of electronic flight bag
By constructing a track model and using adaptive particle swarm algorithm and neural network model, the problem of collaborative modeling and real-time performance of multi-dimensional dynamic factors in the existing technology is solved, and high-precision and efficient optimization of real-time route planning are achieved, improving the safety and economicality of route planning.
Patent Information
- Application Number
- CN202510687123.5
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-05-27
- Publication Date
- 2025-08-01
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
In the face of sudden strong convective weather and dynamic airspace limitations, the existing technology lacks the ability to model multi-dimensional dynamic factors. The balance between real-time and global optimal solution search efficiency of route optimization algorithms has not been effectively solved. The collaborative decision-making mechanism between airborne equipment and ground base stations has bottlenecks such as inconsistent data interaction protocols and redundant verification process, resulting in inefficient rapid deployment and verification of dynamic routes.
By obtaining dynamic data of the flight environment, building a track model, using an adaptive particle swarm algorithm to generate an alternative route set that meets the preset safety threshold, and using a neural network model to update it in real time, combining the time cost function and fuel economy indicators for joint optimization to achieve real-time route planning.
It significantly improves the accuracy of track safety assessment, ensures real-time, and at the same time improves the global search capability. The prediction accuracy of dynamic update neural network models continues to improve, realizing rapid route planning under complex meteorological conditions.
Smart Images
Figure CN120403652A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of intelligent perception and control devices, and particularly to a real-time route dynamic planning method, device and equipment for an electronic flight bag. Background Art
[0002] In recent years, with the rapid growth of air transportation volume and the increasing tension of airspace resources, real-time dynamic route planning technology has become the key to improving flight safety and operation efficiency. The current mainstream systems still rely on predefined airways and periodically updated meteorological data. When facing dynamic airspace restrictions such as sudden severe convective weather and temporary activities, there are often problems of response delay and insufficient optimization. In the prior art, the processing of meteorological data and airspace restriction information mostly adopts an independent analysis mode, lacking the ability to co-model multi-dimensional dynamic factors (such as the spatio-temporal evolution of turbulence and fuel consumption fluctuations). At the same time, the balance between the real-time performance of the route optimization algorithm and the search efficiency of the global optimal solution has not been effectively solved. Although some solutions introduce machine learning models, it is difficult to adapt to the dynamic constraint changes of aircraft in complex airspace environments. In addition, the collaborative decision-making mechanism between airborne devices and ground base stations still has bottlenecks such as inconsistent data interaction protocols and redundant verification processes, restricting the rapid deployment and verification efficiency of dynamic routes. Summary of the Invention
[0003] To achieve the above object, the real-time route dynamic planning method for an electronic flight bag provided by the present invention includes the following steps: Obtain dynamic data of the flight environment, where the dynamic data includes a meteorological turbulence intensity distribution map, a set of coordinates of a temporary airspace restriction polygon, and the current fuel consumption rate parameter of the aircraft; Based on the dynamic data, construct a track model, and use an adaptive particle swarm algorithm to generate a set of alternative routes that meet a preset safety threshold; Extract the time cost function and fuel economy information of each flight segment in the set of alternative routes, and calculate the priority data of each alternative route; Generate a prompt signal through the priority data, and determine whether a route change instruction is received. If so, change the alternative route corresponding to the instruction, and generate real-time data including altitude information, speed information, heading angle, and route; Output the real-time data to the electronic flight bag, and the electronic flight bag feeds back the real-time data to the ground base station.
[0004] Further, the step of obtaining dynamic data of the flight environment includes: At the same moment, obtain a set of characteristic points of the turbulence intensity distribution in the real-time meteorological data, and generate the meteorological turbulence intensity distribution map; At the same moment, receive the sequence of vertex coordinates of the temporary airspace restriction polygon pushed by the ground base station, and generate the set of coordinates of the temporary airspace restriction polygon; At the same time, collect the current fuel consumption rate parameter of the aircraft and associate it with the flight altitude layer information to obtain the current fuel consumption rate parameter of the aircraft.
[0005] Further, the steps of constructing a trajectory model based on the dynamic data include: Establish a mixed training data set including historical turbulence distribution data, airspace restriction trajectories, and fuel consumption characteristics, perform spatio-temporal correlation encoding on the mixed training data set, and generate a training sample set with multi-dimensional labels; Construct a neural network model, whose network input layer receives the three-dimensional grid feature vector of the dynamic data, and the middle layer includes an airspace restriction topology analysis module and a turbulence diffusion time series prediction module; Adjust the parameters of the network input layer to calculate the gradient backpropagation based on the trajectory segment conflict cost function, and generate parameter weight values under various flight states; Inject real-time flight verification data stream, perform state quantization on the model output layer through the parameter weight values under various flight states, and dynamically update the confidence interval threshold of the neural network model; Reverse correct the weight matrix of the fully connected layer of the neural network model through the trajectory safety score to generate the trajectory model with adaptive update ability.
[0006] Further, the working method of the trajectory model includes: Convert the meteorological turbulence intensity distribution map into a three-dimensional coordinate system with the aircraft centroid as the origin, divide the grid cells including longitude, latitude, and altitude, and map each data point to the corresponding grid; Perform spatial topology analysis on the temporary airspace restriction polygon coordinate set, identify its outer vertices and establish a three-dimensional buffer zone, and set obstacle avoidance boundary conditions in combination with the aircraft climb rate and turning radius parameters; Calculate the probability value of each trajectory point encountering turbulence in the grid cell, generate a dynamic diffusion path based on the turbulence gradient change rate, and establish a conflict determination rule in combination with the obstacle avoidance boundary conditions of the buffer zone; Mark the energy consumption level of the trajectory segment in the grid cell according to the fuel consumption rate parameter, and superimpose the probability value and the energy consumption level to generate a three-dimensional trajectory conflict heat map; Adopt a random sampling algorithm to traverse the trajectory point sequence of the heat map, and calculate the comprehensive conflict probability score of each trajectory segment that simultaneously meets the safety threshold and fuel constraint.
[0007] Further, the steps of constructing a trajectory model based on the dynamic data and using an adaptive particle swarm algorithm to generate a set of alternative routes that meet the preset safety threshold include: Initialize the position vector of each particle in the particle swarm as the initial waypoint sequence and the velocity vector as the gradient of the heading angle change according to the comprehensive conflict probability score output by the track model. Construct a fitness function that includes a conflict probability constraint term, a fuel consumption penalty term, and a route smoothness index. The weight of the conflict probability constraint term in the fitness function is dynamically adjusted according to the flight phase. Update the inertia weight coefficient based on the distribution density of the optimal solutions during the particle swarm iteration process, and then screen the non-dominated solution set through the elite particles triggered by the flight segment safety threshold. Correct the particle velocity direction. When adjacent particles are detected to be aggregated, inject a Gaussian perturbation factor to break through the local optimum, and output the non-dominated solution set that meets the safety threshold as the alternative route set.
[0008] Further, the steps of extracting the time cost function and fuel economy information of each flight segment in the alternative route set and calculating the priority data of each alternative route include: Calculate the time cost function of each flight segment, including generating a time cost matrix according to the flight segment distance, flight speed parameters, and the predicted delay time caused by airspace restrictions. Extract the fuel economy information and calculate the predicted total fuel consumption of the flight segment based on the current fuel consumption rate and flight segment climb rate parameters. Calculate the priority score of each alternative route through the time cost function and the predicted total fuel consumption value. Generate a priority sorting queue based on the priority score, and bind the dynamic parameters of the first N routes in the queue to the prompt signal.
[0009] Further, the steps of calculating the priority score of each alternative route through the time cost function and the predicted total fuel consumption value include: Convert the altitude information, speed information, and heading angle into the standard protocol format of the aviation data link, and send them to the electronic flight bag display interface through the on-board communication module. Automatically generate a data packet containing real-time route change parameters in the electronic flight bag, and use an encryption verification algorithm to mark the integrity of the data packet and then send it back to the ground base station.
[0010] An electronic flight bag real-time route dynamic planning device proposed by the present invention includes: An acquisition unit for acquiring flight environment dynamic data, where the dynamic data includes a meteorological turbulence intensity distribution map, a temporary airspace restriction polygon coordinate set, and an aircraft current fuel consumption rate parameter. A model unit for constructing a track model based on the dynamic data and generating an alternative route set that meets a preset safety threshold by using an adaptive particle swarm algorithm. A route unit, which is used to extract the time cost function and fuel economy information of each flight segment in the set of alternative routes, and calculate the priority data of each alternative route; A selection unit, which is used to generate a prompt signal through the priority data, and determine whether a route change instruction is received. If so, it changes the alternative route corresponding to the instruction, and generates real-time data including altitude information, speed information, course angle, and route; A signal unit, which is used to output the real-time data to an electronic flight bag, and the electronic flight bag feeds back the real-time data to a ground base station.
[0011] The present invention also provides a computer device, including a memory and a processor. A computer program is stored in the memory, and when the processor executes the computer program, the steps of the above-mentioned real-time route dynamic planning method for an electronic flight bag are implemented.
[0012] Furthermore, an aircraft electronic flight bag is deployed in the computer device, and the steps of the above-mentioned real-time route dynamic planning method for an electronic flight bag are executed by using the electronic flight bag.
[0013] The real-time route dynamic planning method, device, and equipment provided by the present invention have the following beneficial effects: (1) By fusing the turbulence intensity distribution, airspace restriction polygons, and fuel consumption rate parameters to construct a flight track model, and using grid-based spatio-temporal correlation analysis to generate a conflict heat map, the accuracy of flight track safety assessment is significantly improved; (2) The adaptive particle swarm optimization algorithm introduces an elite particle retention mechanism and a Gaussian perturbation factor, which improves the global search ability while ensuring real-time performance; (3) The time cost function and fuel economy index are jointly optimized through a normalized weight model; (4) By dynamically updating the weights of the neural network model through a learning mechanism, the prediction accuracy of the flight track model under complex meteorological conditions is continuously improved. BRIEF DESCRIPTION OF THE DRAWINGS
[0014] Figure 1 is a flowchart of the real-time route dynamic planning method for an electronic flight bag in an embodiment of the present invention; Figure 2 is a structural block diagram of the real-time route dynamic planning device for an electronic flight bag in an embodiment of the present invention; Figure 3 is a structural schematic diagram of a computer device in an embodiment of the present invention.
[0015] The realization, functional characteristics, and advantages of the object of the present invention will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION OF THE EMBODIMENTS
[0016] In order to make the objectives, technical solutions and advantages of the present invention more clear and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention.
[0017] Refer to Figure 1 , which is a schematic flowchart of a method for real-time route dynamic planning of an electronic flight bag proposed by the present invention. The method includes the following steps: S1. Obtain dynamic data of the flight environment. The dynamic data includes a meteorological turbulence intensity distribution map, a set of coordinates of a temporary airspace restriction polygon, and an aircraft's current fuel consumption rate parameter; In step S1, the step of obtaining dynamic data of the flight environment specifically includes: At the same moment, obtain a set of characteristic points of the turbulence intensity distribution in the real-time meteorological data and generate the meteorological turbulence intensity distribution map; At the same moment, receive the sequence of vertex coordinates of the temporary airspace restriction polygon pushed by the ground base station and generate the set of coordinates of the temporary airspace restriction polygon; At the same moment, collect the aircraft's current fuel consumption rate parameter and associate it with the flight altitude layer information to obtain the aircraft's current fuel consumption rate parameter.
[0018] Specifically, scan the front airspace through an airborne millimeter-wave radar, combine with the meteorological cloud map data packet sent by the satellite, identify the turbulence core area, and generate a three-dimensional turbulence intensity distribution map with the aircraft's centroid as the origin and covering a range of 200 miles ahead. The set of characteristic points of the turbulence intensity distribution includes information characteristics such as the aircraft's spatial coordinates, turbulence intensity value, and turbulence diffusion rate. Receive the NOTAM (Notice to Airmen) data packet broadcast by the ground base station through the ACARS data link, parse the sequence of vertex coordinates of the temporary airspace restriction area therein, and map the vertex coordinates to the aircraft's inertial navigation system reference system to generate the set of coordinates of the temporary airspace restriction polygon. Collect the engine fuel injection amount data in real time through a fuel flow meter, combine with the current thrust parameter output by the FMS, calculate the fuel consumption rate per unit time, obtain the current barometric altitude, true airspeed, and outside temperature through an air data computer, and interpolate and calculate the theoretical fuel flow value based on the engine performance envelope data table; obtain the aircraft's current fuel consumption rate parameter.
[0019] S2. Based on the dynamic data, construct a track model, and use an adaptive particle swarm algorithm to generate a set of alternative routes that meet the preset safety threshold; In step S2, the step of constructing a track model based on the dynamic data includes: Establish a mixed training data set including historical turbulence distribution data, airspace restriction trajectories, and fuel consumption characteristics, perform spatio-temporal correlation encoding on the mixed training data set, and generate a training sample set with multi-dimensional labels; Construct a neural network model, whose network input layer receives the three-dimensional grid feature vector of dynamic data, and the intermediate layer includes an airspace restriction topology analysis module and a turbulent diffusion time series prediction module; Adjust the parameters of the network input layer to calculate the gradient backpropagation based on the flight segment conflict cost function, and generate parameter weight values under various flight states; Inject the real-time flight verification data stream, perform state quantization on the model output layer through the parameter weight values under various flight states, and dynamically update the confidence interval threshold of the neural network model; Reverse correct the weight matrix of the fully connected layer of the neural network model through the track safety score, and generate the track model with the ability of adaptive update.
[0020] Specifically, collect historical flight data (time span ≥ 3 years), and the data includes: turbulent distribution data extracted from the historical echo data of meteorological radar, including the three-dimensional coordinates of the turbulent core area and the intensity attenuation curve; airspace restriction trajectories that analyze the spatio-temporal distribution characteristics of the polygon vertices in the NOTAM database; fuel consumption characteristics that associate the fuel flow rate with the flight altitude and speed parameters in different flight phases (climb, cruise, descent). Thus, a mixed training data set of historical turbulent distribution data, airspace restriction trajectories, and fuel consumption characteristics is obtained. Subsequently, Perform the spatio-temporal correlation encoding process of track points. Use the sliding window technique to intercept track segments (window length 5 minutes, step size 30 seconds), perform spatio-temporal marking on the track points within each window, and pass through the time decay factor Assign higher weights to the recent data:
[0021] In the formula, the sliding window represents , several windows , e is the spatio-temporal data, and a training sample set with multi-dimensional labels is generated. The sample set includes conflict probability labels, fuel consumption level labels, and time cost labels.
[0022] Create a neural network model with an input layer and an intermediate layer. The input layer is designed to receive a three-dimensional grid feature vector (dimension: longitude × latitude × height × feature channels), and the feature channels include: turbulent intensity gradient value, airspace restriction area penetration flag bit, fuel consumption rate deviation amount; the intermediate layer is designed in a modular way, including an airspace restriction topology analysis module and a turbulent diffusion time series prediction module. In the airspace restriction topology analysis module, a graph convolutional network (GCN) is used to construct the polygon vertices of the restricted area into a graph structure, and the topological constraints are propagated through the adjacency matrix; in the turbulent diffusion time series prediction module, causal dilated convolution (Dilated CNN) is used to capture spatio-temporal dependencies, and the dilation coefficient increases exponentially (1, 2, 4,...) to expand the receptive field.
[0023] When adjusting the parameters of the network input layer to calculate the gradient backpropagation based on the flight segment conflict cost function and generate the parameter weight values in various flight states, the conflict cost function is designed as follows:
[0024] In the formula, is the dynamic weight (adjusted in real time by the flight phase classifier), is the predicted value of the conflict probability, is the predicted deviation of fuel consumption, is the historical waypoint heading angle.
[0025] In the step of injecting the real-time flight verification data stream and performing state quantization on the model output layer through the parameter weight values in various flight states, and dynamically updating the confidence interval threshold of the neural network model, the real-time flight data stream is injected into the model through Monte Carlo Dropout, and the confidence interval threshold of the output layer is calculated. Based on the aircraft track safety score in the confidence interval threshold, the weight matrix of the fully connected layer of the neural network model is corrected to generate the track model with the ability of adaptive update.
[0026] In an embodiment of step S2, the working method of the track model includes: Converting the meteorological turbulence intensity distribution map into a three-dimensional coordinate system with the aircraft centroid as the origin, dividing the grid cells including longitude, latitude and altitude, and mapping each data point to the corresponding grid; Performing spatial topology analysis on the temporary airspace restriction polygon coordinate set, identifying its peripheral vertices and establishing a three-dimensional buffer zone, and setting the obstacle avoidance boundary conditions in combination with the aircraft climb rate and turning radius parameters; Calculating the probability value of each track point encountering turbulence in the grid cell, generating a dynamic diffusion path based on the turbulence gradient change rate, and establishing a conflict determination rule in combination with the obstacle avoidance boundary conditions of the buffer zone; Marking the energy consumption level of the flight segment in the grid cell according to the fuel consumption rate parameter, and superimposing the probability value and the energy consumption level to generate a three-dimensional track conflict heat map; Using the random sampling algorithm to traverse the track point sequence of the heat map and calculating the comprehensive conflict probability score of each flight segment that simultaneously meets the safety threshold and fuel constraint.
[0027] In this embodiment, converting the meteorological turbulence intensity distribution map into a three-dimensional coordinate system with the aircraft centroid as the origin, using the following coordinate system conversion formula: Assume that the aircraft centroid coordinates are (x0, y0, z0), and the global coordinates of a certain turbulence point are (x g , y g , z g), then the coordinates (x l , y l , z l ) in the local coordinate system are as follows:
[0028] In the formula, R is the radius of the earth; Identify the outer vertices of the temporary airspace restriction polygon through spatial topology analysis, generate a three-dimensional obstacle avoidance buffer in combination with the aircraft performance parameters, extract the convex hull of the polygon vertex set using the Graham scan algorithm, and the computational complexity is O(nlog n ), to ensure real-time performance, and according to the aircraft climb rate r climb (unit: feet per minute) and the turning radius r turn (unit: nautical miles), calculate the buffer extension distance r buffer :
[0029] Use the anisotropic diffusion equation to describe the evolution of turbulence:
[0030] Among them, I(x, y, z, t) is the turbulence intensity, and the diffusion coefficient tensor D is dynamically adjusted by the gradient change rate . Establish a conflict determination rule in combination with the obstacle avoidance boundary conditions of the buffer: for the track point , if any of the following conditions are met, it is marked as a conflict:
[0031] In the step of marking the energy consumption level of the flight segment in the grid cell according to the fuel consumption rate parameter, and generating a three-dimensional flight track conflict heat map by superimposing the probability value and the energy consumption level, mark the energy consumption level of the flight segment according to the fuel consumption rate parameter, and generate a three-dimensional heat map by superimposing the conflict probability. The heat map uses the generation algorithm:
[0032] Among them, the comprehensive risk value of each grid cell is R, the conflict probability weight , the energy consumption level , is the corresponding flight segment.
[0033] In the step of traversing the track point sequence of the heat map using the random sampling algorithm and calculating the comprehensive conflict probability score of each flight segment that simultaneously meets the safety threshold and fuel constraint, use the random sampling algorithm to traverse the heat map track points and calculate the comprehensive score S j .
[0034] Randomly select N = 1000 tracks in the heat map, and each track contains M = 10 waypoints. The scoring function is:
[0035] In the formula, , and is the heading angle of the i-th waypoint, and R i is the comprehensive risk value of the i-th waypoint.
[0036] In another embodiment of step S2, the steps of constructing a trajectory model based on the dynamic data and generating a set of alternative routes that meet the preset safety threshold by using an adaptive particle swarm algorithm include: According to the comprehensive conflict probability score output by the trajectory model, initialize the position vector of each particle in the particle swarm as the initial waypoint sequence, and the velocity vector as the gradient of the heading angle change; Construct a fitness function that includes a conflict probability constraint term, a fuel consumption penalty term, and a route smoothness index, and the weight of the conflict probability constraint term in the fitness function is dynamically adjusted according to the flight phase; Based on the distribution density of the optimal solutions during the particle swarm iteration process, update the inertia weight coefficient, and then screen the non-dominated solution set through the elite particles triggered by the segment safety threshold; Correct the particle velocity direction. When it is detected that adjacent particles gather, inject a Gaussian perturbation factor to break through the local optimum, and output the non-dominated solution set that meets the safety threshold as the set of alternative routes.
[0037] In this embodiment, initialize the particle swarm parameters according to the conflict probability score output by the trajectory model to ensure that the starting point of the algorithm meets the basic safety constraints. Among them, each particle represents a candidate route, encoded as a discrete waypoint sequence:
[0038] where M = 10 is the number of waypoints, and the initial waypoints are randomly selected from the historical safe route library and need to meet:
[0039] For the definition of the velocity vector, the velocity vector represents the gradient of the heading angle change, and the initialization range is:
[0040] Constraint condition: The sudden change of the heading angle between adjacent waypoints does not exceed 30° (to prevent sharp turns).
[0041] In the step where the weight of the conflict probability constraint term in the fitness function is dynamically adjusted according to the flight phase, construct a fitness function F for multi-objective optimization, and the weight parameters are dynamically adjusted according to the flight phase to match the task requirements. The following functional formula is used for dynamic adjustment:
[0042] In the formula, P conflict : The average value of the track conflict probability (obtained from the output of the track model); F fuel : The predicted fuel consumption value (calculated based on the surface integral of the fuel consumption rate); The second-order norm of the course angle change (quantifying the course smoothness), where γ is the course smoothness weight coefficient. The adjustment rules are as follows in the table:
[0043] During the particle swarm iteration process, in updating the inertia weight coefficient based on the distribution density of the optimal solutions, the inertia weight is dynamically adjusted based on the particle swarm distribution density to balance the global search ability. The above distribution density calculation is to statistically calculate the spatial aggregation degree of the top 10% elite particles in terms of fitness value in the particle swarm:
[0044] In the formula, N e is the number of elite particles. The smaller ρ is, the higher the particle aggregation degree indicates.
[0045] The above formula for updating the inertia weight coefficient is:
[0046] Parameter setting: w max = 0.9, w min = 0.4, ρ threshold = 5 nautical miles. In the formula, is the current inertia weight, is the maximum inertia weight, is the minimum inertia weight, t is the current iteration number, is the maximum iteration number, is the particle aggregation degree, represents the aggregation degree threshold.
[0047] Furthermore, the non-dominated solution set is screened by elite particles triggered by the segment safety threshold. It is through the safety threshold-triggered elite retention mechanism, combined with Gaussian perturbation to break through the local optimum. The non-dominated solution screening is to retain the particles that meet the following conditions in each round of iteration: where the Pareto front is determined by fast non-dominated sorting (NSGA-II framework), so as to obtain the non-dominated solution set of elite particles.
[0048] In the step of correcting the particle velocity direction, when it is detected that adjacent particles are aggregated, a Gaussian perturbation factor is injected to break through the local optimum, and the non-dominated solution set that meets the said safety threshold is output as the alternative route set: Calculate the average distance of the particle swarm ,
[0049] When d avg < d critical = 2 nautical miles, Gaussian perturbation is injected into non-elite particles:
[0050] In outputting the non-dominated solution set that meets the said safety threshold as the alternative route set, the route retention rule is adopted:
[0051] S3. Extract the time cost function and fuel economy information of each flight segment in the set of alternative routes, and calculate the priority data of each alternative route. In step S3, calculate the time cost function of each flight segment, including generating a time cost matrix based on the flight segment distance, flight speed parameter, and predicted delay time caused by airspace restrictions. Extract the fuel economy information, and calculate the predicted total fuel consumption of the flight segment based on the current fuel consumption rate and flight segment climb rate parameter. Calculate the priority score of each alternative route through the time cost function and the predicted total fuel consumption value. Generate a priority sorting queue based on the priority score, and bind the dynamic parameters of the first N routes in the queue to the prompt signal.
[0052] In this embodiment, convert the altitude information, speed information, and heading angle into the standard protocol format of the aviation data link, and send them to the electronic flight bag display interface through the on-board communication module. Automatically generate a data packet containing real-time route change parameters in the electronic flight bag, and use an encryption verification algorithm to mark the integrity of the data packet and then send it back to the ground base station.
[0053] S4. Generate a prompt signal through the priority data, and determine whether a route change instruction is received. If so, change the alternative route corresponding to the instruction, and generate real-time data including altitude information, speed information, heading angle, and route. S5. Output the real-time data to the electronic flight bag, and the electronic flight bag feeds back the real-time data to the ground base station.
[0054] Refer to the appendix Figure 2 The following is a structural block diagram of an electronic flight bag real-time route dynamic planning device proposed by the present invention, including: An acquisition unit, configured to acquire dynamic flight environment data, where the dynamic data includes a meteorological turbulence intensity distribution map, a set of coordinates of a temporary airspace restriction polygon, and the current fuel consumption rate parameter of the aircraft. A model unit, configured to construct a flight path model based on the dynamic data, and use an adaptive particle swarm optimization algorithm to generate a set of alternative routes that meet a preset safety threshold. A route unit, configured to extract the time cost function and fuel economy information of each flight segment in the set of alternative routes, and calculate the priority data of each alternative route. A selection unit, configured to generate a prompt signal through the priority data, and determine whether a route change instruction is received. If so, change the alternative route corresponding to the instruction, and generate real-time data including altitude information, speed information, heading angle, and route. A signal unit for outputting the real-time data to an electronic flight bag, and the electronic flight bag feeds back the real-time data to a ground base station.
[0055] Referring to Figure 3 , an embodiment of the present invention further provides a computer device, which may be a server, and its internal structure may be as Figure 3 shown. The computer device includes a processor, a memory, a display screen, an input device, a network interface, and a database connected through a system bus. Among them, the processor of the computer design 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, a computer program, and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store the corresponding data in this embodiment. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the above method is implemented.
[0056] Those skilled in the art can understand that Figure 3 the structure shown in
[0057] is only a block diagram of a part of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied.
[0058] In summary, Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing 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 methods. Among them, any reference to a memory, storage, database, or other medium provided by the present invention and used in the embodiments can include non-volatile and / or volatile memories. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM, etc.
[0059] It should be noted that, in this document, the terms "including", "comprising", or any other variant thereof are intended to cover non-exclusive inclusion, such that a process, apparatus, article, or method that includes a series of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, apparatus, article, or method. Without further limitation, an element defined by the phrase "including a..." does not exclude the existence of additional identical elements in the process, apparatus, article, or method that includes the element.
[0060] The above are only the preferred embodiments of the present invention, and do not limit the patent scope of the present invention accordingly. Any equivalent structure or equivalent process transformation made by using the specification and drawings of the present invention, or directly or indirectly applied in other related technical fields, shall equally be included in the patent protection scope of the present invention.
Claims
1. A real-time route dynamic planning method for an electronic flight bag, characterized in that, Including the following steps: Obtain dynamic flight environment data, where the dynamic data includes a meteorological turbulence intensity distribution map, a set of coordinates of a temporary airspace restriction polygon, and the current fuel consumption rate parameter of the aircraft; Based on the dynamic data, construct a flight path model, and use an adaptive particle swarm optimization algorithm to generate a set of alternative flight routes that meet the preset safety threshold; Extract the time cost function and fuel economy information of each flight segment in the set of alternative flight routes, and calculate the priority data of each alternative flight route; Generate a prompt signal through the priority data, and determine whether a flight route change instruction is received. If so, change the alternative flight route corresponding to the instruction, and generate real-time data including altitude information, speed information, heading angle, and flight route; Output the real-time data to the electronic flight bag, and the electronic flight bag feeds back the real-time data to the ground base station.
2. The real-time route dynamic planning method for an electronic flight bag according to claim 1, characterized in that The step of obtaining dynamic flight environment data includes: At the same moment, obtain the set of characteristic points of the turbulence intensity distribution in the real-time meteorological data, and generate the meteorological turbulence intensity distribution map; At the same moment, receive the sequence of vertex coordinates of the temporary airspace restriction polygon pushed by the ground base station, and generate the set of coordinates of the temporary airspace restriction polygon; At the same moment, collect the current fuel consumption rate parameter of the aircraft and associate it with the flight altitude layer information to obtain the current fuel consumption rate parameter of the aircraft.
3. The real-time route dynamic planning method of the electronic flight bag according to claim 1, wherein, The step of constructing a flight path model based on the dynamic data includes: Establish a mixed training data set including historical turbulence distribution data, airspace restriction trajectories, and fuel consumption characteristics, perform spatio-temporal correlation encoding on the flight path points of the mixed training data set, and generate a training sample set with multi-dimensional labels; Construct a neural network model, whose network input layer receives the three-dimensional grid feature vector of the dynamic data, and the middle layer includes an airspace restriction topology analysis module and a turbulence diffusion time series prediction module; Adjust the parameters of the network input layer to calculate the gradient backpropagation based on the flight segment conflict cost function, and generate the parameter weight values in various flight states; Inject real-time flight verification data stream, perform state quantization on the model output layer through the parameter weight values in various flight states, and dynamically update the confidence interval threshold of the neural network model; Reverse correct the weight matrix of the fully connected layer of the neural network model through the flight path safety score, and generate the flight path model with adaptive update ability.
4. The real-time route dynamic planning method for an electronic flight bag according to claim 3, wherein: The working method of the flight path model includes: Convert the meteorological turbulence intensity distribution map into a three-dimensional coordinate system with the aircraft centroid as the origin, divide the grid cells including longitude, latitude, and altitude, and map each data point to the corresponding grid; Perform spatial topology analysis on the set of coordinates of the temporary airspace restriction polygon, identify its peripheral vertices and establish a three-dimensional buffer zone, and set obstacle avoidance boundary conditions in combination with the aircraft climb rate and turning radius parameters; Calculate the probability value of each flight path point encountering turbulence in the grid cell, generate a dynamic diffusion path based on the turbulence gradient change rate, and establish a conflict determination rule in combination with the obstacle avoidance boundary conditions of the buffer zone; Mark the energy consumption level of the flight segment in the grid cell according to the fuel consumption rate parameter, and superimpose the probability value and the energy consumption level to generate a three-dimensional flight path conflict heat map; The random sampling algorithm is used to traverse the track point sequence of the heat map, and the comprehensive conflict probability score that satisfies both the safety threshold and fuel constraints for each flight segment is calculated.
5. The real-time route dynamic planning method for an electronic flight bag according to claim 4, characterized in that The steps of constructing a track model based on the dynamic data and using an adaptive particle swarm algorithm to generate a set of alternative routes that meet the preset safety threshold include: According to the comprehensive conflict probability score output by the track model, the position vector of each particle in the particle swarm is initialized as the initial waypoint sequence, and the velocity vector is the gradient of the heading angle change. A fitness function including a conflict probability constraint term, a fuel consumption penalty term, and a route smoothness index is constructed, and the weight of the conflict probability constraint term in the fitness function is dynamically adjusted according to the flight phase. Based on the distribution density of the optimal solution during the particle swarm iteration, the inertia weight coefficient is updated, and then the non-dominated solution set is screened by elite particles triggered by the flight segment safety threshold. The direction of the particle velocity is corrected. When adjacent particles are detected to gather, a Gaussian perturbation factor is injected to break through the local optimum, and the non-dominated solution set that meets the safety threshold is output as the set of alternative routes.
6. The real-time route dynamic planning method for an electronic flight bag according to claim 1, wherein The steps of extracting the time cost function and fuel economy information of each flight segment in the set of alternative routes and calculating the priority data of each alternative route include: Calculating the time cost function of each flight segment, including generating a time cost matrix according to the flight segment distance, flight speed parameters, and the expected delay time caused by airspace restrictions. Extracting the fuel economy information and calculating the predicted total fuel consumption of the flight segment based on the current fuel consumption rate and the flight segment climb rate parameter. Calculating the priority score of each alternative route through the time cost function and the predicted total fuel consumption value. Generating a priority sorting queue based on the priority score, and binding the dynamic parameters of the first N routes in the queue to the prompt signal.
7. The real-time route dynamic planning method for an electronic flight bag according to claim 6, wherein The steps of calculating the priority score of each alternative route through the time cost function and the predicted total fuel consumption value include: Converting the altitude information, speed information, and heading angle into the standard protocol format of the aviation data link, and sending them to the electronic flight bag display interface through the on-board communication module. Automatically generating a data packet containing real-time route change parameters in the electronic flight bag, and using an encryption verification algorithm to mark the integrity of the data packet and then transmitting it back to the ground base station.
8. An electronic flight bag real-time route dynamic planning device, characterized in that, Including: An acquisition unit for acquiring dynamic flight environment data, where the dynamic data includes a meteorological turbulence intensity distribution map, a set of coordinates of a temporary airspace restriction polygon, and the current fuel consumption rate parameter of the aircraft. A model unit for constructing a track model based on the dynamic data and using an adaptive particle swarm algorithm to generate a set of alternative routes that meet the preset safety threshold. A route unit for extracting the time cost function and fuel economy information of each flight segment in the set of alternative routes and calculating the priority data of each alternative route. A selection unit for generating a prompt signal through the priority data, and judging whether a route change instruction is received. If so, changing the alternative route corresponding to the instruction, and generating real-time data including altitude information, speed information, heading angle, and route. A signal unit for outputting the real-time data to the electronic flight bag, and the electronic flight bag feeds back the real-time data to the ground base station.
9. A computer device, comprising a memory and a processor, wherein a computer program is stored in the memory, characterized in that, When the processor executes the computer program, the steps of the real-time route dynamic planning method of the electronic flight bag according to any one of claims 1 to 7 are implemented.
10. The computer device according to claim 1, characterized in that, Deploy an aircraft electronic flight bag in a computer device, and use the electronic flight bag to execute the steps of the real-time route dynamic planning method of the electronic flight bag according to any one of claims 1-7.