Machine learning oriented aircraft operation trajectory accurate segmentation evaluation method and device

Abstract

The application discloses a kind of machine learning-oriented aircraft operation trajectory precision segmentation evaluation method, device, equipment and storage medium.The method comprises: collecting flight trajectory data from flight log or sensor, according to flight task, trajectory is divided into four stages of take-off, cruise, combat and landing, and the evaluation index of each stage is calculated respectively.The application can accurately evaluate the quality of flight trajectory by comprehensively considering the performance of aircraft in different flight stages, especially the influence of key indicators such as speed, acceleration, energy consumption and flight stability in each stage, which is beneficial to extract high-quality flight trajectory as training sample of machine learning model, thereby improving the prediction and optimization ability of model in actual flight task.

Description

Technical Field

[0001] This invention relates to the field of aerospace technology, specifically to a method, apparatus, device, and storage medium for precise segmented evaluation of aircraft trajectories based on machine learning. Background Technology

[0002] The flight trajectories of unmanned aerial vehicles (UAVs) during missions contain a wealth of dynamic information. Evaluating these trajectories can help identify the best-performing flight paths and use them as training samples for machine learning models, thereby optimizing model performance. Furthermore, with the rapid development of UAVs and autonomous flight systems, accurate evaluation and analysis of aircraft trajectories have become increasingly important.

[0003] Chinese patent "A Flight Performance Testing System and Method for Fixed-Wing Unmanned Aerial Vehicles" (Publication No.: CN108045596A, Publication Date: May 18, 2018) records and evaluates the flight trajectory of fixed-wing unmanned aerial vehicles (UAVs) through a flight trajectory evaluation system, and evaluates the performance of UAVs based on the detection parameters. Chinese patent "A Method for Distributed Multi-UAV Relay Network Coverage" (Publication No.: CN116017479A, Publication Date: April 25, 2023) establishes a multi-UAV communication channel model, energy consumption model, and collaborative trajectory optimization mathematical model. It builds a multi-agent deep reinforcement learning algorithm with centralized training and distributed execution, equipping each UAV with an actor and an evaluator. The actor outputs the flight trajectory based on the UAV's state information, and the evaluator outputs the corresponding action value to train the actor. A dual-datastream structure for the actor is constructed to improve the dynamics of the UAVs.

[0004] However, traditional flight trajectory evaluation methods often rely on simple statistical analysis, lack detailed consideration of the characteristics of different flight phases, and cannot make full use of data to optimize the training process of machine learning models.

[0005] Therefore, there is an urgent need for a segmented and precise evaluation method for aircraft flight trajectories, which can provide detailed evaluation of different operational phases of an aircraft and support flight trajectory optimization and machine learning model training. Summary of the Invention

[0006] Purpose of the invention: The present invention aims to provide a method and apparatus for precise segmentation evaluation of aircraft flight trajectories for machine learning, which can perform detailed evaluations of different flight phases and select high-quality trajectories as training samples for machine learning.

[0007] Technical solution: To solve the above technical problems, the technical solution adopted by the present invention is as follows:

[0008] Firstly, a machine learning-based method for accurate segmented evaluation of aircraft trajectories is provided, comprising the following steps:

[0009] (1) Acquire flight trajectory data;

[0010] (2) Segment the flight path;

[0011] (3) Calculate the evaluation index for each segment and complete the segmented evaluation of the flight trajectory.

[0012] As a further aspect of the present invention, the flight trajectory is divided into four segments according to the flight mission: takeoff segment, cruise segment, combat segment, and landing segment.

[0013] As a further aspect of the present invention, the method for calculating the evaluation index of the takeoff phase is as follows:

[0014] S takeoff = k a • A takeoff + k θ •SD( θ takeoff )

[0015] in, S takeoff As an evaluation indicator for the takeoff phase, A takeoff It is takeoff acceleration, SD ( θ takeoff () is the standard deviation of the pitch angle during takeoff. k a and k θ They are A takeoff and SD ( θ takeoff The weighting coefficients of )

[0016] The formulas for calculating takeoff acceleration and the standard deviation of the takeoff pitch angle are as follows:

[0017] ,

[0018] Where ΔV is the velocity difference between the start and end trajectory points of the takeoff segment, and Δt is the time difference between the start and end trajectory points of the takeoff segment. It is the takeoff stage i Pitch angle of a trajectory point, It is the average pitch angle during takeoff. n 1 represents the number of trajectory points during the takeoff phase.

[0019] As a further aspect of the present invention, the method for calculating the evaluation index of the cruise segment is as follows:

[0020] S cruise = k f •F + k h •SD( H )

[0021] in, S cruise The evaluation indicators for the cruise segment. F It refers to energy efficiency, SD ( H ) is the standard deviation of the cruise flight altitude. k f and k h They are F and SD ( H The weighting coefficients of )

[0022] The formulas for calculating energy efficiency and the standard deviation of cruising altitude are as follows:

[0023] ,

[0024] Among them, F T It is the total energy consumption during the cruise phase, F D It is the cruising flight distance. It is the first cruise segment i The aircraft altitude at the two trajectory points, It is the average flight altitude during the cruise phase. n 2 It represents the number of trajectory points during the cruise segment.

[0025] As a further aspect of the present invention, the method for calculating the evaluation index of the combat phase is as follows:

[0026] S combat = k m • M + k r • R

[0027] in, S combat As an evaluation indicator for the combat phase, M For mobility, R For reaction rate, k m and k r They are MThe weighting coefficients of R;

[0028] The formulas for calculating mobility and reaction speed are as follows:

[0029] ,

[0030] in, It is the combat stage number i 3 The maximum acceleration or deceleration during the next maneuver. It is the combat stage number i 3 The frequency of the next maneuver. k acc and k freq They are separate and The weighting coefficients, n 3 represents the total number of maneuvers during the combat phase. T react,j It is the time of the j-th reaction in the combat phase. m It represents the total number of reactions during the combat phase.

[0031] As a further aspect of the present invention, the method for calculating the evaluation index of the landing phase is as follows:

[0032] S landing = k d • D + k v •SD(V landing )

[0033] in, S landing As an evaluation metric for the landing phase, D It refers to landing accuracy, SD(V) landing ) is the standard deviation of the descent speed. k d and k v They are D and SD(V) landing The weighting coefficients of )

[0034] The formulas for calculating the standard deviation of landing accuracy and landing speed are as follows:

[0035] ,

[0036] in, x act , yact These are the actual landing points. x , y coordinate, x pla , y pla Each of the designated landing points x , y coordinate, It is the landing phase. i The descent speeds at the four trajectory points, It is the average descent speed during the landing phase. n 4 represents the number of trajectory points during the descent phase.

[0037] Secondly, a machine learning-oriented device for precise segmentation and evaluation of aircraft trajectories is also provided, comprising:

[0038] The data acquisition unit is used to acquire flight trajectory data;

[0039] Segmentation units are used to segment the flight trajectory;

[0040] The indicator calculation unit is used to calculate the evaluation indicators for each segment;

[0041] The evaluation unit is used to complete the segmented evaluation of the flight trajectory based on the evaluation indicators of each segment.

[0042] Thirdly, a method for processing aircraft trajectory datasets is also provided. These datasets are used for training models for aircraft trajectory planning based on machine learning. The specific steps of this method are as follows:

[0043] 1) The segmented evaluation method described above is used to evaluate each flight trajectory data in the aircraft operation trajectory dataset in segments;

[0044] 2) Select the trajectory point data corresponding to the evaluation indicators that meet the set requirements within each segment to form a new dataset, which will serve as the training sample for the model of aircraft trajectory planning based on machine learning.

[0045] Fourthly, an electronic device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the above-described segmented evaluation method.

[0046] Fifthly, a computer-readable storage medium is also provided, on which a computer program is stored, which, when executed by a processor, implements the steps of the above-described segmented evaluation method.

[0047] Beneficial Effects: The machine learning-oriented method and apparatus for precise segmented evaluation of aircraft flight trajectories proposed in this invention can provide detailed evaluation indicators for flight trajectories at different stages. These indicators help to filter out high-quality flight data, providing more accurate and reliable training samples for machine learning-based flight path prediction and optimization. Furthermore, the method of this invention can also help flight control systems better understand the performance characteristics of different flight stages, optimizing aircraft design and operation. Attached Figure Description

[0048] Figure 1 This is a flowchart of a method according to an embodiment of the present invention. Detailed Implementation

[0049] To make the content of this invention clearer, the invention will be described in detail below with reference to specific embodiments.

[0050] In this example, it is assumed that the drone underwent standard operations in a test mission, including takeoff, cruise, combat, and landing. Figure 1 As shown, the machine learning-based method for accurate segmented evaluation of aircraft trajectories includes the following steps:

[0051] (1) Read flight trajectory data from flight logs or sensors.

[0052] (2) Based on the flight mission, the flight trajectory is divided into four stages: takeoff, cruise, combat and landing.

[0053] The main flight trajectory data for the four stages of this example are as follows:

[0054] Takeoff phase:

[0055] Speed ​​change: from 0 m / s to 30 m / s, taking 10 seconds

[0056] Pitch angle: increased from 0° to 15°.

[0057] Cruise phase:

[0058] Flight distance: 60km

[0059] High stability: fluctuation ±2m

[0060] Energy consumption: 10 kWh.

[0061] Battle Phase:

[0062] Mobility: Performs three rapid maneuvers with a maximum acceleration of 20 m / s².

[0063] Response speed: The average response time for detecting and responding to a threat is 2 seconds.

[0064] Landing phase:

[0065] Landing accuracy: Deviation from the intended landing point is 3m

[0066] Falling speed fluctuation: from 20m / s to 0m / s, fluctuation ±1m / s.

[0067] The weighting coefficient parameters for this example are set as follows:

[0068] k a =0.5, k θ =0.5, k f =0.3, k h =0.7, k m =0.6, kr =0.4, k d =0.8, k v =0.2.

[0069] (3) Calculate the evaluation indicators for the takeoff phase S takeoff :

[0070] S takeoff = k a • A takeoff + k θ •SD( θ takeoff );

[0071] Takeoff acceleration A takeoff :

[0072] ;

[0073] Takeoff stability SD ( θ takeoff ): Assuming simplification to a change

[0074] SD ( θ takeoff ) = 15° - 0° = 15°;

[0075] Evaluation indicators S takeoff :

[0076] S takeoff = 0.5•3+0.5•15=1.5+7.5=9.

[0077] (4) Calculate the evaluation indicators for the cruise phase S cruise

[0078] S cruise = k f •F + k h •SD( H );

[0079] Energy efficiency F :

[0080] ;

[0081] Highly stable SD ( H ):

[0082] SD ( H ) = 2m;

[0083] Evaluation indicators S cruise calculate:

[0084] S cruise =0.3•0.167+0.7•2=0.05+1.4=1.45.

[0085] (5) Calculate the evaluation indicators for the combat phase. S combat :

[0086] S combat = k m • M + k r • R ;

[0087] flexibility M :

[0088] M=20m / s;

[0089] reaction rate R :

[0090] R=2s;

[0091] Evaluation indicators S combat :

[0092] S combat=0.6•20+0.4•2=12+0.8=12.8.

[0093] (6) Calculate the evaluation indicators for the landing phase. S landing :

[0094] S landing = k d • D + k v •SD(V landing )

[0095] landing accuracy D :

[0096] D=3m;

[0097] Landing stability SD(V) landing ): Assuming simplification to a change

[0098] SD(V landing = 1 m / s;

[0099] Evaluation indicators S landing :

[0100] S landing =0.8•3+0.2•1=2.4+0.2=2.6.

[0101] Evaluation metrics for each stage are calculated. This data can be used to select the best-performing flight trajectories as training samples for the machine learning model, thereby optimizing model performance.

[0102] This invention comprehensively considers the performance of an aircraft in different flight phases, especially the impact of key indicators such as speed, acceleration, energy consumption, and flight stability in each phase. It can accurately evaluate the quality of the flight trajectory, which is beneficial for extracting high-quality flight trajectories as training samples for machine learning models, thereby improving the model's prediction and optimization capabilities in actual flight missions.

[0103] Furthermore, this invention also proposes a machine learning-oriented device for precise segmentation and evaluation of aircraft flight trajectories, comprising:

[0104] The data acquisition unit is used to acquire flight trajectory data;

[0105] Segmentation units are used to segment the flight trajectory;

[0106] The indicator calculation unit is used to calculate the evaluation indicators for each segment;

[0107] The evaluation unit is used to complete the segmented evaluation of the flight trajectory based on the evaluation indicators of each segment.

[0108] The technical solution of the above-mentioned segmented extraction device is the same as that of the aforementioned segmented extraction method, and will not be repeated here.

[0109] Furthermore, this invention also proposes a method for processing aircraft trajectory datasets, wherein the aircraft trajectory datasets are used for training models for aircraft trajectory planning based on machine learning. The specific steps of this method are as follows:

[0110] 1) The segmented evaluation method described above is used to evaluate each flight trajectory data in the aircraft operation trajectory dataset in segments;

[0111] 2) Select the trajectory point data corresponding to the evaluation indicators that meet the set requirements within each segment to form a new dataset, which will serve as the training sample for the model of aircraft trajectory planning based on machine learning.

[0112] Based on the same technical solution, the present invention also proposes a computer-readable storage medium for storing one or more programs, wherein the one or more programs include instructions that, when executed by a computing device, cause the computing device to perform the above-described segmented evaluation method.

[0113] Based on the same technical solution, the present invention also proposes a computing device, including one or more processors, one or more memories, and one or more programs, wherein the one or more programs are stored in the one or more memories and configured to be executed by the one or more processors, and the one or more programs include instructions for performing the above-described segmented evaluation method.

[0114] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0115] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0116] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0117] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0118] The above embodiments are merely illustrative of the technical concept of the present invention and should not be construed as limiting the scope of protection of the present invention. Any modifications made to the technical solutions based on the technical concept proposed in this invention shall fall within the scope of protection of this invention.

Claims

1. A machine learning-based method for precise segmented evaluation of aircraft trajectories, characterized in that, Includes the following steps: (1) Acquire flight trajectory data; (2) Segment the flight path; (3) Calculate the evaluation index for each segment and complete the segmented evaluation of the flight trajectory; Based on the flight mission, the flight path is divided into four segments: takeoff segment, cruise segment, combat segment, and landing segment; The calculation method for the evaluation indicators of the takeoff phase is as follows: S takeoff = k a • A takeoff + k θ •SD( θ takeoff ) in, S takeoff As an evaluation indicator for the takeoff phase, A takeoff It is takeoff acceleration, SD ( θ takeoff () is the standard deviation of the pitch angle during takeoff. k a and k θ They are A takeoff and SD ( θ takeoff The weighting coefficients of ) The formulas for calculating takeoff acceleration and the standard deviation of the takeoff pitch angle are as follows: , Where ΔV is the velocity difference between the start and end trajectory points of the takeoff segment, and Δt is the time difference between the start and end trajectory points of the takeoff segment. It is the takeoff stage i Pitch angle of a trajectory point, It is the average pitch angle during takeoff. n 1 represents the number of trajectory points during the takeoff phase; The calculation method for the evaluation indicators of the combat phase is as follows: S combat = k m • M + k r • R in, S combat As an evaluation indicator for the combat phase, M For mobility, R For reaction rate, k m and k r They are M The weighting coefficients of R; The formulas for calculating mobility and reaction speed are as follows: , , among which, is the combat segment number i 3 The maximum acceleration or deceleration during the next maneuver. It is the combat stage. i 3 The frequency of the next maneuver. k acc and k freq It is separate and The weighting coefficients, n 3 represents the total number of maneuvers during the combat phase. T react,j It is the time of the j-th reaction in the combat phase. m It represents the total number of reactions during the combat phase.

2. The segmented evaluation method according to claim 1, characterized in that, The calculation method for the evaluation indicators of the cruise segment is as follows: S cruise = k f •F + k h •SD( H ) in, S cruise The evaluation indicators for the cruise segment. F It refers to energy efficiency, SD ( H ) is the standard deviation of the cruise flight altitude. k f and k h They are F and SD ( H The weighting coefficients of ) The formulas for calculating energy efficiency and the standard deviation of cruising altitude are as follows: , , of which F T It is the total energy consumption during the cruise phase, F D It is the cruising flight distance. It is the first cruise segment i The aircraft altitude at the two trajectory points, It is the average flight altitude during the cruise phase. n 2 It represents the number of trajectory points during the cruise segment.

3. The segmented evaluation method according to claim 1, characterized in that, The calculation method for the evaluation indicators of the landing phase is as follows: S landing = k d • D + k v •SD(V landing ) in, S landing As an evaluation metric for the landing phase, D It refers to landing accuracy, SD(V) landing ) is the standard deviation of the descent speed. k d and k v They are D and SD(V) landing The weighting coefficients of ) The formulas for calculating the standard deviation of landing accuracy and landing speed are as follows: ,in, x act , y act These are the actual landing points. x , y coordinate, x pla , y pla Each of the designated landing points x , y coordinate, It is the landing phase. i The descent speeds at the four trajectory points It is the average descent speed during the landing phase. n 4 represents the number of trajectory points during the descent phase.

4. An apparatus for applying the machine learning-oriented precise segmentation evaluation method for aircraft trajectory as described in any one of claims 1 to 3, characterized in that, include: The data acquisition unit is used to acquire flight trajectory data; Segmentation units are used to segment the flight trajectory; The indicator calculation unit is used to calculate the evaluation indicators for each segment; The evaluation unit is used to complete the segmented evaluation of the flight trajectory based on the evaluation indicators of each segment.

5. A method for processing an aircraft trajectory dataset, wherein the aircraft trajectory dataset is used for training a model for aircraft trajectory planning based on machine learning, characterized in that, The specific steps of this method are as follows: 1) The segmented evaluation method described in any one of claims 1 to 3 is used to perform segmented evaluation on each flight trajectory data in the aircraft operation trajectory dataset; 2) Select the trajectory point data corresponding to the evaluation indicators that meet the set requirements within each segment to form a new dataset, which will serve as the training sample for the model of aircraft trajectory planning based on machine learning.

6. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of any of the segmented evaluation methods described in claims 1 to 3.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of any of the segmented evaluation methods described in claims 1 to 3.

Images