A multi-rotor unmanned aerial vehicle wind measurement method and device, computer equipment and medium

Abstract

This invention provides a method, apparatus, computer equipment, and medium for wind measurement using a multi-rotor unmanned aerial vehicle (UAV), belonging to the field of UAV wind measurement. The method includes: simultaneously acquiring measurement data from a real-world wind sensor mounted on the multi-rotor UAV, the UAV's flight attitude data and flight data, and measurement data from a true-value wind sensor located in the same flow field; using the real-world wind sensor data, UAV flight attitude data, and flight data as combined inputs, and the true-value wind sensor data as the target output, training a machine learning model until the error between the model output and the true-value data is less than a set threshold; during actual UAV flight, inputting the real-time acquired real-world wind sensor data, flight attitude data, and flight data into the trained machine learning model to obtain the output flow field wind speed and direction data. This effectively reduces measurement deviations caused by rotor turbulence, significantly improving the accuracy and reliability of wind measurement results.

Description

Technical Field

[0001] This invention belongs to the field of wind measurement by unmanned aerial vehicles (UAVs), and specifically relates to a wind measurement method, device, computer equipment, and medium for multi-rotor UAVs. Background Technology

[0002] Multi-rotor drones are controllable, highly maneuverable, and low-cost, capable of penetrating high-temperature, high-humidity, and toxic environments, reaching locations inaccessible to personnel. Therefore, they are ideal vehicles and platforms for boundary layer meteorological observation, especially emergency meteorological observation. Wind speed and direction are two fundamental elements frequently measured in meteorological observations. Utilizing multi-rotor drones for meteorological observation requires equipping them with appropriate wind sensors. However, once a multi-rotor drone is started, the airflow caused by the rotor rotation (rotor turbulence) is constantly present, and its nature is the same as natural wind—air movement. No basic wind sensor can distinguish between rotor turbulence and natural wind, and no basic wind sensor can overcome the influence of rotor turbulence. The measurement results include both the atmospheric background flow field and the drone's rotor turbulence. Moreover, the rotor turbulence constantly changes with the drone's flight attitude. It is impossible to directly measure the "true" atmospheric background flow field information from the flow field around the multi-rotor drone; therefore, rotor turbulence correction is necessary for the wind measurement data.

[0003] The turbulence of single-rotor UAVs (such as unmanned helicopters) is mainly downwash, while the turbulence of multi-rotor UAVs is more complex due to the interaction of the turbulence from each rotor. Rotor turbulence is affected by a variety of factors, including the structural parameters of the multi-rotor UAV (blade size, airfoil, motor KV value, etc.), flight attitude and flight data (heading angle, roll angle, pitch angle, speed, etc.), and the turbulence also varies with spatial position (for example, the size of the turbulence varies greatly at different heights above the wing plane). This makes it almost impossible to accurately characterize the impact of rotor turbulence on wind measurement results using mathematical functions, and also means that existing correction methods (subtracting fixed corrections, calculating corrections using aircraft flight parameters, etc.) can only partially eliminate the influence of turbulence, resulting in deviations in the final wind measurement results. Summary of the Invention

[0004] To address the issue of biased wind measurement results in existing technologies, this invention provides a wind measurement method, apparatus, computer equipment, and medium for multi-rotor unmanned aerial vehicles (UAVs).

[0005] To achieve the above objectives, the present invention provides the following technical solution: A method for wind measurement using a multi-rotor unmanned aerial vehicle (UAV), the method comprising: Simultaneously collect measurement data from the real-value wind sensor mounted on the multi-rotor UAV, the UAV's flight attitude data and flight data, as well as measurement data from the true-value wind sensor located in the same flow field; The measured wind sensor data, UAV flight attitude data and flight data are used as combined inputs, and the true wind sensor data is used as the target output to train the machine learning model until the error between the model output and the true data is less than a set threshold. During actual drone flight, real-time collected measured wind sensor data, flight attitude data, and flight data are input into the trained machine learning model to obtain output flow field wind speed and wind direction data.

[0006] Optionally, the true wind sensor is mounted on a hollow bracket, with its installation height greater than or equal to the vertical dimension of the rotor turbulence, and its horizontal distance from surrounding obstacles greater than or equal to 10 times the height of the obstacles.

[0007] Optionally, the measured wind sensor is installed vertically above the wing plane of the multi-rotor UAV and located on the axis of symmetry of the UAV's center of gravity.

[0008] Optionally, the true wind sensor and the measured wind sensor use the same response rate wind sensor, and the sampling time scale and frequency are kept synchronized through the same source time synchronization.

[0009] Optionally, the UAV and the truth sensor are at the same horizontal level, and the horizontal distance between them is greater than or equal to 5 times the wheelbase of the rotary-wing UAV.

[0010] A wind measurement device for a multi-rotor unmanned aerial vehicle (UAV), the device comprising: The measurement module is used to simultaneously collect measurement data from the real-world wind sensor mounted on the multi-rotor UAV, the UAV's flight attitude data and flight data, as well as measurement data from the true-value wind sensor located in the same flow field. The training module is used to train a machine learning model by taking the measured wind sensor data, UAV flight attitude data and flight data as combined inputs and the true wind sensor data as the target output, until the error between the model output and the true data is less than a set threshold. The wind measurement module is used to input real-time collected wind sensor data, flight attitude data, and flight data into the trained machine learning model during actual drone flight to obtain output flow field wind speed and wind direction data.

[0011] A computer-readable storage medium storing a computer program that, when executed by a processor, implements the aforementioned wind measurement method for a multi-rotor unmanned aerial vehicle.

[0012] A computer device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the aforementioned method for wind measurement by a multi-rotor unmanned aerial vehicle.

[0013] The wind measurement method for multi-rotor UAVs provided by this invention has the following beneficial effects: This invention overcomes the limitations of traditional methods by simultaneously collecting measured wind sensor data from a multi-rotor UAV, UAV flight attitude and flight data, and true wind sensor data from the same flow field. This data is then used to train a machine learning model. Traditional methods suffer from complex interactions between the rotors of a multi-rotor UAV, making it difficult to accurately characterize their impact on wind measurement results using mathematical functions. Existing correction methods can only partially eliminate the influence of disturbances, leading to biased wind measurement results. This new method uses measured wind sensor data, flight attitude, and flight data as combined inputs, and true wind sensor data as the target output to train the model. This allows the model to learn the inherent relationships between data under complex disturbance environments. When the error between the model output and the true data is less than a set threshold, the model demonstrates accurate predictive ability. In actual flight, inputting the real-time collected data into the trained model directly yields wind speed and direction data, effectively reducing measurement bias caused by rotor disturbance interactions and significantly improving the accuracy and reliability of wind measurement results. This provides more precise technical support for wind measurement applications of multi-rotor UAVs in meteorological monitoring, environmental research, and other fields. Attached Figure Description

[0014] To more clearly illustrate the embodiments and design schemes of the present invention, the accompanying drawings required for this embodiment will be briefly described below. The drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0015] Figure 1 This is a flowchart illustrating a wind measurement method for a multi-rotor unmanned aerial vehicle (UAV) according to an exemplary embodiment of the present invention.

[0016] Figure 2 This is a schematic diagram of a data acquisition scenario provided by the present invention according to an exemplary embodiment.

[0017] Figure 3 This is a schematic diagram of model training according to an exemplary embodiment of the present invention.

[0018] Figure 4 This is a schematic diagram illustrating the use of a model according to an exemplary embodiment of the present invention.

[0019] Figure 5 This is a block diagram of a multi-rotor unmanned aerial vehicle (UAV) wind measurement device according to an exemplary embodiment of the present invention. Detailed Implementation

[0020] To enable those skilled in the art to better understand and implement the technical solutions of the present invention, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. The following embodiments are only used to more clearly illustrate the technical solutions of the present invention and should not be construed as limiting the scope of protection of the present invention.

[0021] The technical solutions provided by the various embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0022] First, this invention provides a wind measurement method for multi-rotor UAVs, specifically as follows: Figure 1 As shown, it includes the following steps: S101. Simultaneously collect measurement data from the real-value wind sensor mounted on the multi-rotor UAV, the UAV's flight attitude data and flight data, as well as the measurement data from the true-value wind sensor located in the same flow field.

[0023] The flight attitude data includes pitch angle, roll angle, and yaw angle, while the flight data includes horizontal speed, vertical speed, and motor speed.

[0024] In one embodiment, the true wind sensor is mounted on a perforated bracket. To ensure that the installation height is unaffected by the ground or solid walls, its installation height is greater than or equal to the vertical dimension of the rotor turbulence. Furthermore, meteorological studies indicate that when the horizontal distance exceeds 10 times the height of an obstacle, the impact of the obstacle on the measurement is negligible. Therefore, to reduce the influence of surrounding obstacles on the measurement results, the horizontal distance between the true wind sensor and surrounding obstacles is greater than or equal to 10 times the height of the obstacle. The measured wind sensor is mounted vertically above the wing plane of the multi-rotor UAV, and to ensure that the installation position does not affect the UAV's flight balance, it is located on the axis of symmetry of the UAV's center of gravity. Additionally, the true wind sensor and the measured wind sensor use the same high-response rate wind sensors and maintain sampling timescale and frequency synchronization through co-source time synchronization. Moreover, the UAV and the true sensor are at the same height, and to ensure that the UAV rotor turbulence does not affect the measurement of the true sensor, a certain distance is maintained between them. Fluid dynamics simulations show that the influence of rotor turbulence beyond 5 times the UAV's wheelbase is negligible; therefore, the horizontal distance between them is greater than or equal to 5 times the UAV's wheelbase.

[0025] S102. Using the measured wind sensor data, UAV flight attitude data, and flight data as combined inputs, and the true wind sensor data as the target output, train the machine learning model until the error between the model output and the true wind sensor data is less than a set threshold.

[0026] The machine learning model is an artificial neural network model. The data samples used for training, as described in the previous steps, all come from a uniform natural flow field unaffected by the ground or solid walls. Furthermore, to maintain the model's predictive performance and improve the accuracy of the prediction results, it is necessary to periodically repeat the aforementioned data collection and model training steps, collecting new training data to supplement and validate the machine learning model.

[0027] S103. During the actual flight of the UAV, the real-time collected measured wind sensor data, flight attitude data and flight data are input into the trained machine learning model to obtain the output flow field wind speed and wind direction data.

[0028] In this way, we can obtain the "real" wind speed and direction data of the flow field where the drone is located.

[0029] By employing the aforementioned method, and simultaneously collecting measured wind sensor data from a multi-rotor UAV, along with the UAV's flight attitude and flight data, and ground truth wind sensor data from the same flow field, and using this data to train a machine learning model, the limitations of traditional methods can be overcome. Traditional methods suffer from complex interactions between the rotors of a multi-rotor UAV, making it difficult to accurately characterize their impact on wind measurement results using mathematical functions. Existing correction methods can only partially eliminate the influence of disturbances, leading to biases in the wind measurement results. This new method, however, uses measured wind sensor data, flight attitude, and flight data as combined inputs, and ground truth wind sensor data as the target output to train the model, enabling the model to learn the inherent relationships between data under complex disturbance environments. When the error between the model output and the ground truth data is less than a set threshold, it indicates that the model possesses accurate predictive capabilities. In actual flight, inputting the corresponding real-time collected data into the trained model directly yields flow field wind speed and direction data, effectively reducing measurement bias caused by rotor disturbance interactions, significantly improving the accuracy and reliability of wind measurement results, and providing more precise technical support for wind measurement applications of multi-rotor UAVs in meteorological monitoring, environmental research, and other fields.

[0030] Based on the above steps, the present invention also provides an embodiment.

[0031] like Figure 2 As shown, prepare two high-response-rate wind sensors of the same specifications. Using sensors of identical specifications ensures that their basic measurement performance is consistent. The flow field around the fuselage of a multi-rotor UAV changes rapidly, requiring sensors with high response rates; ultrasonic wind sensors can be used.

[0032] A wind sensor is installed on a hollowed-out bracket in a natural flow field. The installation height and position should be such that the airflow around the sensor is not affected by the ground and surrounding solid walls. It is used to measure the true value of the flow field (i.e., a true value sensor).

[0033] Install another wind sensor (the measured sensor) at an appropriate position above the wing plane of the multi-rotor drone, as high as possible without affecting the aircraft's balance performance.

[0034] To maintain time synchronization between the true sensor and the measured sensor, the sampling frequency must also be kept consistent, through time synchronization from the same source or other means.

[0035] The measurement system for maneuvering a multi-rotor UAV is located at the same altitude as the ground truth sensor, but at an appropriate distance (rotor turbulence does not affect the ground truth sensor).

[0036] Simultaneously acquire data from the ground truth sensor and the multi-rotor UAV measurement system (measured sensor data, multi-rotor UAV flight and attitude data).

[0037] like Figure 3 As shown, the measurement data of the sensor (true value sensor) in the natural flow field is used as the "true value" for training the neural network correction model; the upwind sensor data and aircraft attitude data of the multi-rotor UAV are used as the model input to train the neural network model; when the difference between the model output result and the "true value" reaches the set threshold, the model training ends.

[0038] like Figure 4 As shown, by inputting the measurement data of the multi-rotor UAV and real-time data such as the UAV's attitude into the trained neural network model, the real-time natural flow field information can be obtained after the model is corrected.

[0039] The above steps can be repeated periodically to check the real-time quality of the corrected wind measurement data from the multi-rotor UAV. At the same time, the check data can also be added to the model training samples to supplement the training of the corrected model.

[0040] For multi-rotor drone measurement systems with the same configuration (multi-rotor drones and wind sensors of the same model, with the same connection method and materials used), the correction model can be universal and can be mass-produced.

[0041] In the above embodiments, it should be noted that both the ground truth flow field used to train the model and the multi-rotor UAV measurement samples (wind measurement data, flight attitude data, etc.) should be kept away from the influence of the ground or solid walls (buildings, forests, etc.). That is, the ground truth sensor and the multi-rotor UAV measurement system should be far enough away from the ground and solid walls (such as building walls, mountains, etc.) so that neither of them will be affected by the ground and solid walls. For example, the ground truth sensor can be installed on a hollow tower at a certain height (the installation height should be such that the flow field at the ground truth sensor is not affected by the ground friction, and the degree of hollowing of the hollow tower should have almost no effect on the measurement of the ground truth sensor).

[0042] The ground truth flow field used to train the model and the multi-rotor UAV measurement samples (wind measurement data, flight attitude data, etc.) should come from the same flow field. That is, when the multi-rotor UAV acquires wind measurement data, flight attitude data, etc., it is in the same uniform flow field as the ground truth sensor, at the same height, with the same sampling frequency, and the measurements are performed synchronously.

[0043] The ground truth flow field used to train the model should not interfere with the measurement samples (wind data, flight attitude, etc.) from the multi-rotor UAV. The multi-rotor UAV system should maintain an appropriate distance from the ground truth sensor, such that the disturbance from the multi-rotor UAV does not affect the ground truth sensor, but both should be in the same uniform background flow field.

[0044] Secondly, the present invention also provides a wind measurement device for multi-rotor unmanned aerial vehicles, such as... Figure 5 As shown, it includes: The measurement module 201 is used to simultaneously collect measurement data from the measured wind sensor mounted on the multi-rotor UAV, the flight attitude data and flight data of the UAV, and the measurement data from the true wind sensor located in the same flow field.

[0045] The training module 202 is used to train a machine learning model by taking the measured wind sensor data, UAV flight attitude data and flight data as combined inputs and the ground truth wind sensor data as the target output, until the error between the model output and the ground truth data is less than a set threshold.

[0046] The wind measurement module 203 is used to input the real-time collected measured wind sensor data, flight attitude data and flight data into the trained machine learning model during the actual flight of the UAV, and obtain the output flow field wind speed and wind direction data.

[0047] The present invention also provides a computer-readable storage medium storing a computer program that can be used to execute the above-described... Figure 1 The steps of the provided multi-rotor UAV wind measurement method.

[0048] This invention also provides a computer device. At the hardware level, the computer device includes a processor, an internal bus, a network interface, memory, and non-volatile memory, and may also include other hardware required for various operations. The processor reads the corresponding computer program from the non-volatile memory into memory and then executes it to achieve the above-mentioned functions. Figure 1 The steps of the provided multi-rotor UAV wind measurement method.

[0049] 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.

[0050] 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, as well as 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... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0051] 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.

[0052] These computer program instructions may also be loaded onto a computer or other programmable data processing apparatus to cause a series of operational steps to be performed on the computer or other programmable apparatus to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable apparatus 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.

[0053] It should be noted that the specific embodiments described above enable those skilled in the art to more fully understand the present invention, but do not limit the present invention in any way. Therefore, although the present invention has been described in detail in this specification, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the present invention; and all technical solutions and improvements that do not depart from the spirit and scope of the present invention are covered within the protection scope of the patent of the present invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

Claims

1. A wind measurement method for a multi-rotor unmanned aerial vehicle (UAV), characterized in that, The method includes: Simultaneously collect measurement data from the real-value wind sensor mounted on the multi-rotor UAV, the UAV's flight attitude data and flight data, as well as measurement data from the true-value wind sensor located in the same flow field; The measured wind sensor data, UAV flight attitude data and flight data are used as combined inputs, and the true wind sensor data is used as the target output to train a machine learning model until the error between the model output and the true wind sensor data is less than a set threshold. During actual drone flight, real-time collected wind sensor data, flight attitude data, and flight data are input into the trained machine learning model to obtain output flow field wind speed and wind direction data.

2. The method according to claim 1, characterized in that, The true wind sensor is mounted on a hollow bracket, with its installation height greater than or equal to the vertical dimension of the rotor turbulence, and its horizontal distance from surrounding obstacles greater than or equal to 10 times the height of the obstacles.

3. The method according to claim 1, characterized in that, The measured wind sensor is installed vertically above the wing plane of the multi-rotor UAV and located on the axis of symmetry of the UAV's center of gravity.

4. The method according to claim 1, characterized in that, The true wind sensor and the measured wind sensor use the same response rate wind sensor, and the sampling time scale and frequency are kept synchronized through the same source time synchronization.

5. The method according to claim 1, characterized in that, The drone and the truth sensor are at the same horizontal level, and the horizontal distance between them is greater than or equal to 5 times the wheelbase of the rotorcraft drone.

6. A wind measurement device for a multi-rotor unmanned aerial vehicle (UAV), characterized in that, The device includes: The measurement module is used to simultaneously collect measurement data from the real-value wind sensor mounted on the multi-rotor UAV, the UAV's flight attitude data and flight data, as well as the measurement data from the true-value wind sensor located in the same flow field. The training module is used to train a machine learning model by taking the measured wind sensor data, UAV flight attitude data and flight data as combined inputs and the true wind sensor data as the target output, until the error between the model output and the true data is less than a set threshold. The wind measurement module is used to input real-time collected wind sensor data, flight attitude data, and flight data into the trained machine learning model during actual drone flight to obtain output flow field wind speed and wind direction data.

7. A computer-readable storage medium, characterized in that, The storage medium stores a computer program, which, when executed by a processor, implements the method described in any one of claims 1 to 5.

8. A computer device, characterized in that, The method includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the method described in any one of claims 1 to 5.

Images