Paraglider driving training system based on artificial intelligence
Through the paragliding driving training system based on artificial intelligence, the flight training model and virtual scenes are used to solve the problem of scarce coaching resources in paragliding training, and the rapid and effective skills mastery are achieved, reducing training costs and time.
Patent Information
- Application Number
- CN202411412916.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2024-10-10
- Filing Date
- 2024-10-11
- Publication Date
- 2024-12-10
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Paragliding is limited by the scarce coaching resources, making it difficult for beginners to master driving skills quickly, resulting in high training costs and time.
Develop a paragliding driving training system based on artificial intelligence. By obtaining flight training data, collecting umbrella control data in real time, using pre-trained flight training models to generate virtual flight scenarios and trajectories, providing real-time feedback and auxiliary teaching to help students master driving. Skill.
The cost and time of paragliding training is reduced, and through virtual training scenarios and real-time data collection, effective auxiliary teaching for students is achieved and the learning efficiency of driving skills is improved.
Smart Images
Figure CN119107858A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of artificial intelligence technology, and in particular to a paraglider driving training system based on artificial intelligence. Background Art
[0002] Paragliding is becoming increasingly popular, but it is also an extreme sport that requires athletes to take off, fly and land in strong winds using a paraglider.
[0003] The shape of the paraglider wing is similar to a willow leaf, with an upper wing surface and a lower wing surface. When the wing moves relative to the air, a certain pressure is generated under the action of air pressure to shape the canopy and maintain a certain rigidity to provide lift. Control ropes are connected to the leading edge, trailing edge, left edge and right edge of the paraglider wing. The driver changes the lift by pulling the control ropes to change the shape of the wing.
[0004] Beginners need to learn how to fly a paraglider using a control rope, but paragliding is difficult to popularize due to the scarcity of instructors. Generally, before a beginner can fly a paraglider, they need to undergo several weeks of ground parachute control training, with an instructor by their side to guide and correct the movements.
[0005] In order to enable beginners to quickly master paragliding driving skills through training equipment, thereby reducing the cost and time of paragliding training, the present application provides a paragliding driving training system based on artificial intelligence. Summary of the invention
[0006] In order to overcome the problems existing in the related art, the present application provides a paraglider driving training system based on artificial intelligence, comprising:
[0007] S1. Obtain flight training data;
[0008] S2. demonstrating a virtual flight training scene according to the flight training data;
[0009] S3, collecting parachute control data in real time; the parachute control data is data obtained by encoding the electrical signal of the paraglider simulation controller;
[0010] S4, determining a flight training result according to the flight parachute control data, the flight airflow data and the flight training model;
[0011] Wherein, the flight training model is used to predict the flight trajectory of the paraglider controlled by the flight parachute control data;
[0012] The pre-trained flight training model is installed on the paragliding training equipment, and the paragliding training equipment is used to display the flight training results to the trainee according to the output of the flight training model;
[0013] The flight training model is trained by a paraglider flight control training set, and the paraglider flight control training set includes airflow data, paraglider attitude data and the actual flight trajectory of the paraglider.
[0014] In one embodiment, before S1, the method further includes:
[0015] N flight training data are constructed according to N actual flight scenarios, and the flight training data include flight training items, flight training instructions and flight training objectives.
[0016] In one implementation, S2 specifically includes:
[0017] Get flight training programs;
[0018] Searching and loading virtual scene data in device storage according to the flight training project;
[0019] Generate a virtual flight training scene according to the virtual scene data; the flight training scene includes a virtual glider, a virtual three-dimensional training ground and the flight training instructions;
[0020] According to the flight training instruction, corresponding instruction marks are displayed in the flight training scene, and the instruction marks are glider driving operation steps.
[0021] In one embodiment, the paraglider simulation controller is a simulation control rope, the simulation control rope is electrically connected to the paraglider training equipment, and the simulation control rope includes 4;
[0022] S3 specifically includes:
[0023] The electrical signal of the simulated control rope is collected in real time; the electrical signal of the simulated control rope is the stroke, force and trigger time of the simulated control rope;
[0024] The electrical signal of the simulated control rope is encoded into the flight parachute control data with a timestamp, and the flight parachute control data is V1=[S, α, C 1t ,k] t .
[0025] In one implementation, S4 specifically includes:
[0026] Acquire the flight parachute control data and flight airflow data; wherein the flight parachute control data includes canopy area, angle of attack, lift coefficient and inclination angle, and the flight airflow data includes airflow direction and airflow speed;
[0027] Encoding the flight parachute control signal and the flight airflow data and inputting them into the flight training model;
[0028] Outputting a virtual flight trajectory through the flight training model;
[0029] The flight training result is determined according to the virtual flight trajectory and the flight training target.
[0030] In one embodiment, after S4, the method further includes:
[0031] Output paraglider attitude correction data based on flight training results;
[0032] An attitude correction action is determined according to the paraglider attitude correction data.
[0033] In one embodiment, before S4, the method further includes:
[0034] Acquire a flight control training set; the flight control training set includes training airflow data, training parachute control data, and a real flight trajectory;
[0035] The training parachute control data is collected through a real glider flight process;
[0036] Constructing the flight training model;
[0037] The paraglider training model is trained using the flight control training set.
[0038] In one embodiment, the MSE loss function of the flight training model is defined as:
[0039]
[0040] Where N is the size of the data set, i represents the data number in the data set, r(t) is the predicted glider flight trajectory, r'(t) is the actual glider flight trajectory, λ is the regularization weight, and W n is the nth network hidden layer.
[0041] In one embodiment, further comprising: the PDF loss function of the flight training model is:
[0042]
[0043] in, Represents the prediction error at time t, N is the size of the data set, and i represents the data number in the data set.
[0044] In one embodiment, the physical equation for the trajectory r(t) is:
[0045]
[0046] v 3t =v2t +v 1t
[0047] x t =x0+v 3tx dt
[0048] y t =y0+v 3ty dt
[0049] Among them, x t is the horizontal lateral displacement of the paraglider, y t is the horizontal longitudinal displacement of the paraglider, z t is the height of the paraglider, z0 is the initial height of the paraglider, x0 is the initial lateral coordinate of the paraglider, y0 is the initial longitudinal coordinate of the paraglider, t is the time, v 3tx dt is the horizontal displacement increment at time t, v 3ty dt is the horizontal longitudinal displacement increment of the paraglider at time t, v 1t is the wind speed at time t, v 2t is the glider airspeed at time t, v 3t is the ground speed of the glider at time t, and S is the wing area of the glider.
[0050] The technical solution provided by this application may have the following beneficial effects:
[0051] After the model is trained, the trainee selects flight training data and then performs simulation training on the human-computer interaction device. The model is fed with data through a 4-DOF control rope, and then the model predicts the flight trajectory of the glider in real time based on the trainee's operation. The embodiment of the present application can achieve end-to-end output of virtual paragliding flight scenes through a pre-trained flight training model, and provide auxiliary teaching to paragliding trainees through human-computer interaction.
[0052] It should be understood that the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present application. BRIEF DESCRIPTION OF THE DRAWINGS
[0053] The above and other objects, features and advantages of the present application will become more apparent through a more detailed description of exemplary embodiments of the present application in conjunction with the accompanying drawings, wherein the same reference numerals generally represent the same components in the exemplary embodiments of the present application.
[0054] Figure 1 It is a schematic diagram of the process of the paraglider driving training system shown in the embodiment of the present application;
[0055] Figure 2 It is another flow chart of the paraglider driving training system shown in the embodiment of the present application;
[0056] Figure 3 This is another flow chart of the paraglider driving training system shown in the embodiment of the present application. DETAILED DESCRIPTION
[0057] The preferred embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although the preferred embodiments of the present application are shown in the accompanying drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments described herein. On the contrary, these embodiments are provided to make the present application more thorough and complete, and to fully convey the scope of the present application to those skilled in the art.
[0058] The terms used in this application are for the purpose of describing specific embodiments only and are not intended to limit this application. The singular forms of "a", "said" and "the" used in this application and the appended claims are also intended to include plural forms unless the context clearly indicates other meanings. It should also be understood that the term "and / or" used herein refers to and includes any or all possible combinations of one or more associated listed items.
[0059] It should be understood that although the terms "first", "second", "third", etc. may be used in this application to describe various information, this information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of this application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Thus, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of this application, the meaning of "multiple" is two or more, unless otherwise clearly and specifically defined.
[0060] In order to reduce the cost and time of paragliding training, the paragliding driving training system shown in the embodiment of the present application is installed on a human-computer interaction device, and helps trainees master paragliding driving skills by constructing a virtual training scene.
[0061] The present application embodiment provides a paraglider driving training system based on artificial intelligence, such as Figure 1 As shown, the following steps are included:
[0062] S1. Obtain flight training data;
[0063] Specifically, the flight training data is constructed according to an actual flight scenario, and the flight training data includes flight training items, flight training instructions, and flight training objectives.
[0064] Optionally, the flight training data may be set according to the piloting skills required for the paragliding certificate level test.
[0065] For example, if the flight training data is divided into five levels, A, B, C, D, and E, according to the level of paragliding certificates, then in the A-level flight training data, the flight training item is ground-skimming flight, the flight training goal is forward takeoff, and the flight training instruction is a takeoff operation instruction.
[0066] Specifically, the flight training data is pre-loaded in the human-computer interaction device. When the trainee logs in for training, the flight training data can be loaded according to the trainee's choice.
[0067] S2. demonstrating a virtual flight training scene according to the flight training data;
[0068] Further, step S2 is as follows: Figure 2 As shown, including:
[0069] S201. Obtain flight training programs;
[0070] S202, searching and loading virtual scene data in device storage according to the flight training project;
[0071] S203, generating a virtual flight training scene according to the virtual scene data;
[0072] S204: Display a corresponding indication mark in the flight training scene according to the flight training indication.
[0073] Specific, virtual flight training scenes are displayed on the display screen.
[0074] Optionally, the display screen may be a display output of VR glasses, a television, a mobile phone, or other human-computer interaction devices.
[0075] Exemplarily, if the selected flight training data is level A, the generated virtual flight training scene is a hillside terrain, the flight training instructions are instructions for take-off operations, and the flight training goal is to fly a virtual glider to take off on a hillside terrain.
[0076] In an embodiment of the present application, trainees can perform simulation training according to the takeoff operation instructions on the display screen.
[0077] S3, real-time collection of flight parachute control data;
[0078] Specifically, step S3 includes the following steps:
[0079] The electrical signal of the simulated control rope is collected in real time; the electrical signal of the simulated control rope is the stroke, force and trigger time of the simulated control rope;
[0080] The electrical signal of the simulated control rope is encoded into the flight parachute control data with a timestamp, and the flight parachute control data is V1=[α, k, S, C1] t .
[0081] In gliding flight, the paraglider has four control ropes, which are connected to the leading edge, trailing edge, left edge and right edge of the canopy. Adjusting the angle of the leading edge and trailing edge can control the flying speed of the paraglider, and adjusting the angle of the left edge and right edge can control the turning of the paraglider.
[0082] Therefore, the paraglider training device described in the embodiment of the present application is provided with four simulated control ropes, namely, simulated control ropes. The simulated control ropes are electrically connected to the control host, and when the trainee pulls the simulated control ropes, the control host obtains the flight parachute control data sent by the pulled simulated control ropes. In the traditional computer algorithms based on physical engines, physical algorithms or physical models, in order to simulate the flight trajectory of the paraglider, real-time calculations based on input signals and physical formulas are required, which places high demands on the computing power of the device.
[0083] The paraglider driving training system provided in the embodiment of the present application performs end-to-end output through a flight training model, which is a PINN neural network model.
[0084] In the end-to-end output, the flight paraglider control data is directly input into the flight training model, and the flight training model directly predicts and outputs the flight trajectory of the paraglider.
[0085] S4, determining a flight training result according to the flight parachute control data, airflow data and a flight training model;
[0086] Furthermore, S4 Figure 3 As shown, the following steps are included:
[0087] S401, obtain the flight parachute control data V1 = [α, k, S, C1] t and flight air flow data
[0088] A2=(q,θ) t ;
[0089] S402, encoding the flight parachute control signal and the flight airflow data and inputting them into the flight training model;
[0090] S403, outputting a virtual flight trajectory through the flight training model;
[0091] S404: Determine the flight training result according to the virtual flight trajectory and the flight training target.
[0092] Specifically, the flight parachute control data V1 = [α, k, S, C1]t The flight parachute control data include canopy area S, angle of attack α, lift coefficient C1 and inclination angle k.
[0093] Specifically, the flight airflow data includes airflow direction and airflow speed.
[0094] Specifically, the flight training model is trained by a flight control training set, and the paraglider attitude training set includes training airflow data A2=(q,θ) t And training parachute data V2 = [α, k, S, C1] t Among them, q is the air velocity, θ is the air direction, S is the canopy area, α is the angle of attack, k is the inclination angle, C 1t is the lift coefficient.
[0095] Specifically, the flight training model is a PINN physical model.
[0096] In the embodiment of the present application, the MSE loss function of the flight training model is defined as
[0097]
[0098] Where N is the size of the data set, i represents the data number in the data set, r(t) is the predicted glider flight trajectory, r'(t) is the actual glider flight trajectory, λ is the regularization weight, and W n is the nth network hidden layer.
[0099] In an embodiment of the present application, in order to prevent the model from overfitting, each layer of the loss function is constrained using regularization.
[0100] Furthermore, in order to make the paraglider flight trajectory predicted by the flight training model conform to the laws of physics, in the embodiment of the present application, the paraglider flight trajectory is defined as a three-dimensional coordinate sequence with a timestamp, and the paraglider flight trajectory is r(t)=(x t ,y t ,z t ).
[0101] Specifically, the PDF loss function of the flight training model is:
[0102]
[0103] in, Represents the prediction error at time t, N is the size of the data set, and i represents the data number in the data set.
[0104] The shape of the paraglider wing is similar to a willow leaf, divided into an upper wing surface and a lower wing surface. When the wing moves relative to the air, a certain pressure is generated under the action of air pressure to shape the canopy and maintain a certain rigidity to provide lift.
[0105] Control ropes are connected to the leading edge, trailing edge, left edge and right edge of the paraglider wing. The pilot changes the lift by pulling the control ropes to change the shape of the wing.
[0106] Therefore, the virtual flight trajectory predicted by the flight training model needs to comply with the following physical equations to ensure that the predicted virtual flight trajectory complies with the actual paragliding flight laws.
[0107] Among them, the physical equation of the trajectory r(t) conforms to the following formula:
[0108]
[0109] v 3t =v 2t +v 1t
[0110] x t =x0+v 3tx dt
[0111] y t =y0+v 3ty dt
[0112] Among them, x t is the horizontal lateral displacement of the paraglider, y t is the horizontal longitudinal displacement of the paraglider, z t is the height of the paraglider, z0 is the initial height of the paraglider, x0 is the initial lateral coordinate of the paraglider, y0 is the initial longitudinal coordinate of the paraglider, t is the time, v 3tx dt is the horizontal displacement increment at time t, v 3ty dt is the horizontal longitudinal displacement increment of the paraglider at time t, v 1t is the wind speed at time t, v 2t is the glider airspeed at time t, v 3t is the ground speed of the glider at time t, and S is the wing area of the glider.
[0113] Specifically, the magnitude of the paraglider lift is proportional to the area of the canopy and is closely related to the relative angle between the wing surface and the air, that is, the lift coefficient of the glider is related to the glider's angle of attack, and the glider's angle of attack changes over time, causing the lift coefficient to change.
[0114] Therefore, it is also necessary to define the expression of the lift coefficient of the glider at time t as:
[0115]
[0116] Among them, C1 is the glider lift coefficient at time t, C0 is the initial glider lift coefficient, and α is the glider attack angle.
[0117] Specifically, the initial glider lift coefficient is the lift coefficient after the canopy of the glider is fully deployed.
[0118] When training the model, the collected glider data set is input for training, and the MSE loss function and physical loss function are used to enable the model to make predictions based on the physical model. Thus, end-to-end prediction output can be achieved during training.
[0119] After the model is trained, the trainee selects flight training data, and then performs simulation training on the human-computer interaction device, inputs data into the model through a 4-degree-of-freedom control rope, and then the model predicts the flight trajectory of the glider in real time based on the trainee's operation. The embodiment of the present application can achieve end-to-end output of virtual paragliding flight scenes through a pre-trained flight training model, and provide auxiliary teaching to paragliding trainees through human-computer interaction.
[0120] Regarding the device in the above embodiment, the specific way in which each module performs the operation has been described in detail in the embodiment of the method, and will not be elaborated here.
[0121] The scheme of the present application has been described in detail above with reference to the accompanying drawings. In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments. Those skilled in the art should also be aware that the actions and modules involved in the description are not necessarily required for the present application.
[0122] In addition, it can be understood that the steps in the method of the embodiment of the present application can be adjusted in order, combined and deleted according to actual needs, and the modules in the device of the embodiment of the present application can be combined, divided and deleted according to actual needs.
[0123] In addition, the method according to the present application may also be implemented as a computer program or a computer program product, which includes computer program code instructions for executing some or all of the steps in the above method of the present application.
[0124] Alternatively, the present application can also be implemented as a non-temporary machine-readable storage medium (or computer-readable storage medium, or machine-readable storage medium) on which executable code (or computer program, or computer instruction code) is stored. When the executable code (or computer program, or computer instruction code) is executed by a processor of an electronic device (or electronic device, server, etc.), the processor executes part or all of the steps of the above-mentioned method according to the present application.
[0125] Those skilled in the art will further appreciate that the various illustrative logical blocks, modules, circuits, and algorithm steps described in connection with the application herein may be implemented as electronic hardware, computer software, or combinations of both.
[0126] The flow chart and block diagram in the accompanying drawings show the possible architecture, function and operation of the system and method according to multiple embodiments of the present application. In this regard, each square box in the flow chart or block diagram can represent a part of a module, a program segment or a code, and the part of the module, the program segment or the code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the square box can also occur in a sequence different from that marked in the accompanying drawings. For example, two continuous square boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each square box in the block diagram and / or the flow chart, and the combination of the square boxes in the block diagram and / or the flow chart can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.
[0127] The embodiments of the present application have been described above, and the above description is exemplary, not exhaustive, and is not limited to the disclosed embodiments. Many modifications and changes will be apparent to those of ordinary skill in the art without departing from the scope and spirit of the described embodiments. The selection of terms used herein is intended to best explain the principles of the embodiments, practical applications, or improvements to the technology in the market, or to enable other persons of ordinary skill in the art to understand the embodiments disclosed herein.
Claims
1. A paraglider driving training system based on artificial intelligence, characterized in that: include: S1. Obtain flight training data; S2. demonstrating a virtual flight training scene according to the flight training data; S3, collecting parachute control data in real time; the parachute control data is data obtained by encoding the electrical signal of the paraglider simulation controller; S4, determining a flight training result according to the flight parachute control data, the flight airflow data and the flight training model; Wherein, the flight training model is used to predict the flight trajectory of the paraglider controlled by the flight parachute control data; The pre-trained flight training model is installed on the paragliding training equipment, and the paragliding training equipment is used to display the flight training results to the trainee according to the output of the flight training model; The flight training model is trained by a paraglider flight control training set, and the paraglider flight control training set includes airflow data, paraglider attitude data and the actual flight trajectory of the paraglider.
2. The artificial intelligence-based paraglider driving training system according to claim 1, characterized in that: Prior to S1, this also included: N flight training data are constructed according to N actual flight scenarios, and the flight training data include flight training items, flight training instructions and flight training objectives.
3. The artificial intelligence-based paraglider driving training system according to claim 2, characterized in that: S2 specifically includes: Get flight training programs; Searching and loading virtual scene data in device storage according to the flight training project; Generate a virtual flight training scene according to the virtual scene data; the flight training scene includes a virtual glider, a virtual three-dimensional training ground and the flight training instructions; According to the flight training instruction, corresponding instruction marks are displayed in the flight training scene, and the instruction marks are glider driving operation steps.
4. The artificial intelligence-based paraglider driving training system according to claim 1, characterized in that: The paraglider simulation controller is a simulation control rope, the simulation control rope is electrically connected to the paraglider training equipment, and the simulation control rope includes 4; S3 specifically includes: The electrical signal of the simulated control rope is collected in real time; the electrical signal of the simulated control rope is the stroke, force and trigger time of the simulated control rope; The electrical signal of the simulated control rope is encoded into the flight parachute control data with a timestamp, and the flight parachute control data is V1=[S, α, C 1t ,k] t .
5. The artificial intelligence-based paraglider driving training system according to claim 1, characterized in that: S4 specifically includes: Acquire the flight parachute control data and flight airflow data; wherein the flight parachute control data includes canopy area, angle of attack, lift coefficient and inclination angle, and the flight airflow data includes airflow direction and airflow speed; Encoding the flight parachute control signal and the flight airflow data and inputting them into the flight training model; Outputting a virtual flight trajectory through the flight training model; The flight training result is determined according to the virtual flight trajectory and the flight training target.
6. The artificial intelligence-based paraglider driving training system according to claim 1, characterized in that: After S4, it also includes: Output paraglider attitude correction data based on flight training results; An attitude correction action is determined according to the paraglider attitude correction data.
7. The artificial intelligence-based paraglider driving training system according to claim 1, characterized in that: Prior to S4, this also included: Acquire a flight control training set; the flight control training set includes training airflow data, training parachute control data, and a real flight trajectory; The training parachute control data is collected through a real glider flight process; Constructing the flight training model; The paraglider training model is trained using the flight control training set.
8. The artificial intelligence-based paraglider driving training system according to claim 7, characterized in that: Also includes: The MSE loss function of the flight training model is defined as Where N is the size of the data set, i represents the data number in the data set, r(t) is the predicted glider flight trajectory, r'(t) is the actual glider flight trajectory, λ is the regularization weight, and W n is the nth network hidden layer.
9. The artificial intelligence-based paraglider driving training system according to claim 8, characterized in that: Also includes: The PDF loss function of the flight training model is: in, Represents the prediction error at time t, N is the size of the data set, and i represents the data number in the data set.
10. The artificial intelligence-based paraglider driving training system according to claim 9, characterized in that: The physical equation of the trajectory r(t) is: v 3t =v 2t +v 1t x t =x0+v 3tx dt y t =y0+v 3ty dt Among them, x t is the horizontal lateral displacement of the paraglider, y t is the horizontal longitudinal displacement of the paraglider, z t is the height of the paraglider, z0 is the initial height of the paraglider, x0 is the initial lateral coordinate of the paraglider, y0 is the initial longitudinal coordinate of the paraglider, t is the time, v 3tx dt is the horizontal displacement increment at time t, v 3ty dt is the horizontal longitudinal displacement increment of the paraglider at time t, v 1t is the wind speed at time t, v 2t is the glider airspeed at time t, v 3t is the ground speed of the glider at time t, and S is the wing area of the glider.