Machine learning-oriented aircraft moving trajectory fine segmentation extraction method and device

Abstract

The invention discloses a machine learning-oriented aircraft running trajectory refined segmented extraction method, device and equipment and a storage medium. The method comprises the following steps: acquiring parameters such as speed, acceleration and course angle of an aircraft from a flight log; calculating the maximum value of a single section of track; calculating the trajectory variation of each trajectory point; calculating a track change threshold value; and judging abrupt change points and carrying out refined segmentation. According to the method provided by the invention, changes of the speed, the acceleration and the course angle are comprehensively considered, the state abrupt change points in the flight path can be accurately identified, and segmentation extraction is automatically completed. According to the method, the precision of flight path analysis is effectively improved, high-quality data support is provided for machine learning model training, and meanwhile, the requirements of different flight tasks are met.

Description

Technical Field

[0001] The present invention relates to the field of aircraft trajectory analysis, and in particular to a method and device for extracting aircraft trajectory fine segments for machine learning. Background Art

[0002] When an aircraft is performing a mission, its trajectory contains rich dynamic information. Effective segmentation extraction of trajectory data can help analyze the flight status of the aircraft, optimize flight path planning, and provide higher-quality training data for machine learning models.

[0003] In existing research, the Chinese patent "A method and system for air-ground collaborative inspection" (publication number: CN115146882A, publication date: September 6, 2022) automatically generates the shortest path according to several monitoring areas and divides the shortest path according to the UAV's cruising range, realizing the automatic monitoring flight trajectory segmentation after the addition and deletion of the monitoring area, and can dynamically and reasonably divide it according to the effective mileage of the UAV, so that the air-ground collaborative inspection plan can be automatically planned and executed; the Chinese invention application "A method, device, equipment and storage medium for determining the flight trajectory" (publication number: CN118089718 A, Publication Date: May 28, 2024), determine the intermediate flight trajectory from the planned flight trajectory set, convert the intermediate flight trajectory into the three-dimensional Cartesian coordinate system, and obtain the target flight trajectory; China invention application "Method, system, electronic device and storage medium for determining the flight trajectory of an unmanned aerial vehicle" (Publication No.: CN116382347A, Publication Date: July 4, 2023), based on the ground coordinates of each vertex of the obstacle and the ground coordinates of each vertex of the target area, use the coordinate transfer matrix to determine the relative coordinates of the position points at each moment of the preset flight trajectory to calculate the robustness score and thus determine the flight trajectory.

[0004] However, the segmentation methods for aircraft trajectories in the prior art mostly rely on rough preset rules, which cannot effectively reflect the subtle changes in the state during the flight process, resulting in insufficient segmentation accuracy and cannot well meet the needs of machine learning training data.

[0005] Therefore, there is an urgent need for a refined segmentation extraction method that can automatically identify mutation points based on the changes in speed, acceleration and heading angle in the aircraft trajectory to provide support for subsequent flight trajectory optimization and machine learning model training. Summary of the invention

[0006] Purpose of the invention: The purpose of the present invention is to provide a method, device, equipment and storage medium for fine-grained segmentation extraction of aircraft trajectory for machine learning, which can automatically identify mutation points in the aircraft trajectory and perform fine segmentation by reading parameters such as speed, acceleration and heading angle from the flight log.

[0007] Technical solution: To solve the above technical problems, the technical solution adopted by the present invention is: In a first aspect, a method for extracting fine segmentation of aircraft trajectory for machine learning is provided, the method comprising the following steps: (1) Obtain flight trajectory data; (2) The mutation point is determined according to the set mutation point rules, and the mutation point is used as the segmentation point to realize the segmented extraction of the flight trajectory data.

[0008] As a further solution of the present invention, the flight trajectory data at least includes the flight speed, acceleration and heading angle of each trajectory point.

[0009] As a further solution of the present invention, all trajectory points in the flight trajectory data are traversed in chronological order, and the mutation point is determined according to the following mutation point setting rule: When the trajectory point i When the distance from the previous mutation point is greater than or equal to the maximum value of a single segment trajectory, the trajectory point is determined i is the mutation point; When the trajectory point i The distance from the previous mutation point is less than the maximum value of a single segment trajectory, and the trajectory point i When the trajectory change is greater than or equal to the trajectory change threshold, the trajectory point is determined i is the mutation point; Otherwise, the trajectory point i Not a mutation point.

[0010] As a further solution of the present invention, the maximum value of a single-segment trajectory is set to:

[0011] in, L max is the maximum value of a single-segment trajectory, V avg is the average speed in the flight trajectory data, k V is a coefficient related to the flight environment and mission requirements, and [·] indicates rounding.

[0012] As a further solution of the present invention, the trajectory change calculation formula of the trajectory point is:

[0013] in, J ( i ) is the trajectory point i The trajectory change, ΔV i , ΔA i and Δθ i Represents the trajectory pointsi The change in velocity, acceleration and heading angle, w 1. w 2 and w 3 are ΔV i , ΔA i and Δθ i The weight coefficient of .

[0014] As a further solution of the present invention, the trajectory change threshold is set as:

[0015] in, J threshold is the trajectory change threshold, σ V , σ A , σ θ are the standard deviations of flight speed, acceleration and heading angle, respectively. c 1. c 2. c 3 are σ V , σ A , σ θ The weight coefficient of .

[0016] In a second aspect, a device for extracting fine segmentation of aircraft trajectory for machine learning is also provided, comprising: A data acquisition unit, used for acquiring flight trajectory data from a flight log; A mutation point determination unit, used for determining mutation points in flight trajectory data; The segmented extraction unit is used to extract the flight trajectory data in segments by taking the mutation points as segmented points.

[0017] In a third aspect, a method for processing an aircraft trajectory data set is also provided, wherein the aircraft trajectory data set is used for model training for aircraft trajectory planning based on machine learning, and the specific steps of the method are as follows: 1) Using the segmented extraction method described above, segmented extraction is performed on each aircraft trajectory in the aircraft trajectory data set; 2) The segmented trajectories extracted from step 1) constitute a new data set as training samples for the model of aircraft trajectory planning based on machine learning.

[0018] In a fourth aspect, an electronic device is also provided, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above-mentioned segmented extraction method when executing the computer program.

[0019] In a fifth aspect, a computer-readable storage medium is also provided, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the above-mentioned segmented extraction method are implemented.

[0020] Beneficial effects: The method, device, equipment and storage medium for fine-grained segmentation extraction of aircraft trajectory for machine learning provided by the present invention can automatically calculate the trajectory change at each moment by obtaining data such as the aircraft's speed, acceleration and heading angle from the flight log, and judge the aircraft's mutation point based on the standard deviation and threshold to perform fine segmentation extraction. The method is highly adaptable and can adjust the trajectory segmentation strategy according to the needs of different flight missions to meet the requirements of a variety of flight environments and aircraft types. Compared with the prior art, the present invention can more accurately capture the mutation points of the aircraft's state during the execution of the mission, thereby making the segmentation results more accurate and representative, and at the same time can provide higher quality training data for the machine learning model, thereby improving the model's accuracy in aircraft path planning and dynamic change prediction. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 It is a method flow chart of an embodiment of the present invention. DETAILED DESCRIPTION

[0022] In order to make the content of the present invention clearer, the present invention is described in detail below in conjunction with specific embodiments.

[0023] like Figure 1 As shown, the present invention proposes a method for extracting fine segmentation of aircraft trajectory for machine learning, the method comprising the following steps: Step 1: Read flight trajectory data from the flight log, including but not limited to: speed V : Indicates the speed of the aircraft relative to the ground; Acceleration A : Indicates the instantaneous acceleration change of the aircraft; Heading angle θ: represents the angular change of the aircraft relative to the ground reference direction; Other optional parameters include altitude, aircraft attitude angle, etc., but the above three parameters are the main parameters for calculating trajectory changes in this method.

[0024] In this embodiment, the flight trajectory data of a certain aircraft in a mission as shown in Table 1 (sampled once per second) is read from the flight log: speed V (Unit: m / s), acceleration A (Unit: m / s 2 ) and heading angle θ (unit: °).

[0025] Table 1

[0026] Step 2: Calculate the maximum value of a single segment trajectory L max : .

[0027] In this embodiment, the average speed in the flight trajectory data is calculated V avg : .

[0028] Assume coefficients related to flight environment and mission requirements k V =10, the maximum value of a single segment trajectory L max for: .

[0029] Step 3: Calculate the trajectory change of each trajectory point:

[0030] in, J ( i ) is the trajectory point i The trajectory change, ΔV i , ΔA i and Δθ i Represents the trajectory points i The change in velocity, acceleration and heading angle, w 1. w 2 and w 3 are ΔV i , ΔA i and Δθ i The weight coefficient of .

[0031] , ,

[0032] Among them, V i , A i and θ i Represents the trajectory points i The velocity, acceleration and heading angle, V i-1 , A i-1 and θ i-1 Represents the trajectory points i -1 for velocity, acceleration and heading.

[0033] In this embodiment, it is assumed that the weight coefficient w 1=0.5, w2=1, w 3=0.1, and the trajectory change is calculated as shown in Table 2: Table 2

[0034] Step 4: Calculate the trajectory change threshold for the aircraft operation J threshold :

[0035]

[0036]

[0037]

[0038] in, c 1. c 2. c 3 is the weight coefficient. σ V , σ A , σ θ are the standard deviations of velocity, acceleration and heading angle, respectively, are the average values ​​of velocity, acceleration and heading angle in the time window or distance window, n is the number of data points in the window.

[0039] It should be noted here that the time window T window The size of the distance window is set according to the dynamic performance of the aircraft, such as 1 second, 5 seconds or 10 seconds. D window It is 100 meters, 500 meters or set according to the flight mission requirements of the aircraft, and the standard deviation can be calculated using a sliding window method to more smoothly determine the mutation points of the flight status.

[0040] In this embodiment, the calculated speed, acceleration and heading angle standard deviations are:

[0041]

[0042]

[0043] Assuming weight coefficient c 1=0.1, c 2=4, c 3=0.05, calculated J threshold : .

[0044] Step 5: Traverse all trajectory points in the flight trajectory data in chronological order, determine the mutation point according to the following mutation point rules, and use the mutation point as the segmentation point to achieve segmented extraction of the flight trajectory data. The mutation point rules include: When the trajectory point i When the distance from the previous mutation point is greater than or equal to the maximum value of a single segment trajectory, the trajectory point is determined i is the mutation point; When the trajectory point i The distance from the previous mutation point is less than the maximum value of a single segment trajectory, and the trajectory point i When the trajectory change is greater than or equal to the trajectory change threshold, the trajectory point is determined i is the mutation point; Otherwise, the trajectory point i Not a mutation point.

[0045] In this embodiment, the step of determining the mutation point includes: According to the estimation, the trajectory point i The maximum distance from the previous mutation point is approximately: .

[0046] Therefore, the trajectory points in this embodiment i The distance from the previous mutation point is less than the maximum value of the single segment trajectory. L max , so we need to continue according to J threshold Make further judgment.

[0047] Step 1: J (2)=1.15, less than J threshold , continue to judge J (3) It should be noted here that since the judgment of the mutation point is compared with the previous point, the first trajectory point does not need to be judged.

[0048] Step 2: J (3)=1.85, greater than J threshold ,therefore J (3) is the mutation point, set J (3) is the segment breakpoint.

[0049] Step 3: J (4)=1.8, greater than J threshold ,therefore J (4) is the mutation point, set J (4) is the segment breakpoint.

[0050] Step 4:J (5)=1.7, greater than J threshold ,therefore J (5) is the mutation point, set J (5) is the segment breakpoint.

[0051] Step 5: J (6)=1.4, less than J threshold , continue to judge.

[0052] The method provided by the present invention comprehensively considers the changes in speed, acceleration and heading angle, can accurately identify the state mutation points in the flight trajectory, and automatically complete segment extraction. This method effectively improves the accuracy of flight trajectory analysis, provides high-quality data support for machine learning model training, and adapts to the needs of different flight missions.

[0053] Furthermore, the present invention also proposes a device for extracting fine segmentation of aircraft trajectory for machine learning, comprising: A data acquisition unit, used for acquiring flight trajectory data from a flight log; A mutation point determination unit, used for determining mutation points in flight trajectory data; The segmented extraction unit is used to extract the flight trajectory data in segments by taking the mutation points as segmented points.

[0054] The technical solution of the above-mentioned segmented extraction device is consistent with the technical solution of the aforementioned segmented extraction method, and will not be repeated here.

[0055] Furthermore, the present invention also proposes a method for processing an aircraft trajectory data set, wherein the aircraft trajectory data set is used for model training for aircraft trajectory planning based on machine learning, and the specific steps of the method are as follows: 1) Using the segmented extraction method described above, segmented extraction is performed on each aircraft trajectory in the aircraft trajectory data set; 2) The segmented trajectories extracted from step 1) constitute a new data set as training samples for the model of aircraft trajectory planning based on machine learning.

[0056] Based on the same technical solution, the present invention also proposes a computer-readable storage medium storing one or more programs, wherein the one or more programs include instructions, and when the instructions are executed by a computing device, the computing device executes the above-mentioned segment extraction method.

[0057] Based on the same technical solution, the present invention also proposes a computing device, comprising 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 are configured to be executed by the one or more processors, and the one or more programs include instructions for executing the above-mentioned segmentation extraction method.

[0058] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Furthermore, the present invention may take the form of a computer program product implemented 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.

[0059] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0060] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0061] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

Claims

1. A fine segmentation extraction method for aircraft trajectory based on machine learning, characterized in that: The method comprises the following steps: (1) Obtain flight trajectory data; (2) The mutation point is determined according to the set mutation point rules, and the mutation point is used as the segmentation point to realize the segmented extraction of the flight trajectory data.

2. The method according to claim 1, characterized in that: The flight trajectory data at least includes the flight speed, acceleration and heading angle of each trajectory point.

3. The method according to claim 1, characterized in that Traverse all trajectory points in the flight trajectory data in chronological order and determine the mutation point according to the following mutation point setting rules: When the trajectory point i When the distance from the previous mutation point is greater than or equal to the maximum value of a single segment trajectory, the trajectory point is determined i is the mutation point; When the trajectory point i The distance from the previous mutation point is less than the maximum value of a single segment trajectory, and the trajectory point i When the trajectory change is greater than or equal to the trajectory change threshold, the trajectory point is determined i is the mutation point; Otherwise, the trajectory point i Not a mutation point.

4. The method according to claim 3, characterized in that: The maximum value of a single-segment trajectory is set to: ,in, L max is the maximum value of a single-segment trajectory, V avg is the average speed in the flight trajectory data, k V is a coefficient related to the flight environment and mission requirements, and [·] indicates rounding.

5. The method according to claim 3, characterized in that: The calculation formula of the trajectory change of the trajectory point is: , where ΔV i , ΔA i and Δθ i Represents the trajectory points i The change in velocity, acceleration and heading angle, w 1. w 2 and w 3 are ΔV i , ΔA i and Δθ i The weight coefficient of .

6. The method according to claim 3, characterized in that: The trajectory change threshold is set as: ,in, J threshold is the trajectory change threshold, σ V , σ A , σ θ are the standard deviations of flight speed, acceleration and heading angle, respectively. c 1. c 2. c 3 are σ V , σ A , σ θ The weight coefficient of .

7. A fine segmentation extraction device for aircraft trajectory for machine learning, characterized in that: include: A data acquisition unit, used for acquiring flight trajectory data from a flight log; A mutation point determination unit, used for determining mutation points in flight trajectory data; The segmented extraction unit is used to extract the flight trajectory data in segments by taking the mutation points as segmented points.

8. A method for processing an aircraft trajectory data set, wherein the aircraft trajectory data set is used for model training for aircraft trajectory planning based on machine learning, characterized in that: The specific steps of this method are as follows: 1) Using the segmented extraction method as described in any one of claims 1 to 6 to segmentally extract each aircraft trajectory in the aircraft trajectory data set; 2) The segmented trajectories extracted from step 1) constitute a new data set as training samples for the model of aircraft trajectory planning based on machine learning.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the segment extraction method according to any one of claims 1 to 6 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the segment extraction method according to any one of claims 1 to 6 are implemented.

Images