Unmanned aerial vehicle safe take-off and landing control method under strong wind conditions

Abstract

The application discloses a UAV safe take-off and landing control method under strong wind conditions and relates to the field of UAVs. The UAV safe take-off and landing control method under strong wind conditions comprises the following steps: pre-inspection evaluation is carried out before take-off, take-off evaluation values are comprehensively obtained, it is determined whether take-off conditions are met, if yes, the UAV is started, and the main motor power is adjusted according to the take-off evaluation values; wind field estimation values are output and calibrated, and each flight parameter of the UAV is pre-adjusted; it is checked in real time whether the UAV ascending protection mechanism is triggered, if yes, the pre-adjusted flight parameters are modified, the take-off hovering height is dynamically adjusted, the motor power proportion is distributed in real time, and the UAV is flown steadily to the take-off hovering height; pre-inspection evaluation is carried out before the UAV lands, landing evaluation values are obtained, it is determined whether landing is selected based on the landing evaluation values, if yes, the wind speed near the ground is collected at the deceleration point to determine the wind field type, the landing mode is adaptively matched, the UAV is landed steadily, and the UAV take-off and landing control task is completed.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicles (UAVs), and more particularly to a method for controlling the safe take-off and landing of UAVs under strong wind conditions. Background Technology

[0002] Currently, the takeoff and landing control of drones first checks whether all sensors are normal. After confirming that the parameters are correct, a takeoff command can be sent. After the command is triggered, the drone first starts the motors in a specific order and with a specific power. Based on the sensor data, the speed of each motor is calculated and fine-tuned in real time to resist wind interference and ensure a smooth takeoff.

[0003] After the drone completes its mission and returns to the take-off and landing area, it first enters a pre-landing state at a certain altitude above the ground. It controls the motors to slowly reduce their speed while monitoring the altitude above the ground in real time. If it detects that the drone's altitude is deviated due to ground turbulence, it controls and fine-tunes the rotor thrust to ensure that the landing speed is uniformly reduced. When the drone detects that the descent speed has been reduced to zero and the altitude is close to the ground, it cuts off the motor power to shut down the motors and completes the drone landing process.

[0004] For example, the invention patent with announcement number CN112537444B provides a method for automatic wind alignment during hovering of a compound-wing UAV. When the UAV takes off and lands vertically in rotor mode, it automatically aligns its nose with the oncoming wind direction in two phases: a hovering phase and a heading adjustment phase, thus improving the UAV's wind resistance. When the UAV has just taken off or is preparing for vertical landing, it first enters the hovering phase. After a hovering time t, it immediately enters the heading adjustment phase. Based on roll angle commands, and combined with the horizontal component of the rotor thrust, the rotor rotational torque difference, and the fixed-wing rudder, the UAV's yaw rate control capability is improved to adjust its heading, achieving automatic wind alignment.

[0005] Based on the above technical solutions, it was found that because a fixed-point hovering adjustment mode is set during the take-off and landing phase, the algorithm limits the hovering delay and threshold for the drone. Under the complex environmental interference of strong wind conditions, once there is a rapid change in wind direction and wind force, the drone in the hovering state is constrained by the fixed time delay and cannot respond to the change immediately. The change in wind field interferes with the position and attitude of the hovering drone, causing the motor power consumption required to correct the deviation to exceed the physical limit of the drone, which poses certain safety hazards.

[0006] Automatic wind alignment relies on calculations of wind field conditions. However, wind field conditions change drastically under gusts, and the estimated wind speed vector may lag significantly behind the actual situation. The delay in control commands issued by the UAV causes abnormal deviations in the UAV's attitude due to gust interference. To return to normal flight, the UAV needs to make significant adjustments to its attitude angles, generating additional control inputs. This can lead to control input conflicts or saturation within the control system, causing overload of the UAV's attitude control, resulting in violent oscillations, disrupting the preset UAV upwind operation, and causing it to fall into an unstable state of repeated adjustments. Summary of the Invention

[0007] This invention provides a safe take-off and landing control method for drones under strong wind conditions, which solves the safety problems caused by the inability of drones in fixed hovering mode to respond to environmental changes in a timely manner under complex strong wind interference, as well as the overload of drone control, in the existing technology, and improves the flexibility and safety of drone take-off and landing under strong wind conditions.

[0008] To achieve the above-mentioned objectives, the present invention provides the following technical solution:

[0009] A method for safe takeoff and landing control of unmanned aerial vehicles (UAVs) under strong wind conditions is provided. This method includes: pre-takeoff assessment of the wind field environment and the UAV's own status to obtain a comprehensive takeoff assessment value, determining whether the takeoff conditions are met; if so, starting the UAV with the initial set power and adjusting the main motor power according to the takeoff assessment value; inputting real-time wind field environmental data into a wind field prediction model to output a wind field prediction value; calibrating the wind field prediction value based on the real-time wind field status and pre-adjusting various flight parameters of the UAV; during flight, real-time verification of whether the UAV triggers the ascent protection mechanism; if triggered, correcting the pre-adjusted UAV flight parameters, dynamically adjusting the takeoff hovering altitude, and real-time allocating motor power ratios until stable flight reaches the takeoff hovering altitude; pre-landing assessment of the UAV to obtain a landing assessment value, determining whether to land based on the landing assessment value; if landing, collecting near-ground wind speed at the deceleration point to determine the wind field type, and adaptively matching the landing mode according to the wind field type to ensure a smooth landing and complete the UAV takeoff and landing control task.

[0010] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following:

[0011] 1. This invention provides a safe take-off and landing control method for unmanned aerial vehicles (UAVs) under strong wind conditions. Before take-off, a pre-assessment of the UAV is performed to obtain a comprehensive take-off assessment value, determining whether the take-off conditions are met. This prevents damage to the UAV due to excessively harsh wind conditions exceeding its safety limits. The power of the main motor is adjusted based on the take-off assessment value. Real-time wind field environmental data is input into a wind field prediction model, outputting a wind field prediction value. The wind field prediction value is calibrated based on the real-time wind field status, and various flight parameters of the UAV are pre-adjusted. This allows the UAV to call upon the pre-adjusted flight parameters in real time during flight, shortening the time required for flight state transitions. The system minimizes response delays to ensure the drone's flight stability. During flight, it continuously checks whether the drone has triggered the ascent protection mechanism. If triggered, it corrects the pre-adjusted flight parameters and dynamically adjusts the takeoff and hovering altitude, allocating motor power ratios in real time to ensure the drone can smoothly fly to the takeoff and hovering altitude, preventing the drone from tipping over due to exceeding physical limits. Before landing, the system performs a pre-assessment to obtain the drone's landing assessment value. Based on the landing assessment value, it determines whether to land. If landing is selected, it collects near-ground wind speed at the deceleration point to determine the wind field type and adaptively matches the landing mode according to the wind field type, enabling the drone to land smoothly and completing the drone takeoff and landing control task.

[0012] 2. This invention assesses whether the UAV can meet flight conditions by using takeoff evaluation values ​​based on the wind field environment and the UAV's own status before takeoff. It comprehensively quantifies the performance of the UAV and the quality of the environment, identifies high-risk environments and its own parameters in advance, and eliminates potential accidents from the source. When the takeoff conditions are within the safe threshold for UAV takeoff, the current wind field environment and the UAV's own status are considered to meet the takeoff conditions, and the UAV can be launched. Once the safe threshold for takeoff is exceeded, the UAV returns a feedback signal to the control terminal indicating that the takeoff conditions are not met, and waits until the takeoff conditions are met before taking off, ensuring that the takeoff process can be completed under safe conditions.

[0013] 3. Compared with existing technologies, this solution dynamically adjusts the hovering altitude of the UAV during takeoff based on the actual flight status and environmental conditions under complex environmental interference in strong winds. This ensures flexible adjustment of the UAV's flight parameters in response to changes in wind force and speed, thereby improving the safety factor during flight. By adjusting the UAV's flight parameters through both advance prediction and real-time correction, this solution ensures the UAV's response speed in complex environments and makes specific adjustments to real-time changes, maintaining a relatively stable flight state. This solves the problem of violent oscillations caused by increased control input due to drastic changes in wind conditions, ensuring stable operation during takeoff and landing and improving flight safety during sudden changes in wind fields. Attached Figure Description

[0014] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying 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 of a method for controlling the safe take-off and landing of unmanned aerial vehicles (UAVs) under strong wind conditions, provided in an embodiment of the present invention.

[0016] Figure 2 This is the ground effect curve stored in the UAV take-off and landing control system provided in this embodiment of the invention. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the described embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0018] Unless otherwise defined, the technical or scientific terms used in this invention shall have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms “first,” “second,” and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Similarly, the terms “an,” “a,” or “the,” and similar terms do not indicate a quantity limitation, but rather indicate the presence of at least one. The terms “comprising,” “including,” or “including,” and similar terms mean that the element or object preceding the word encompasses the element or object listed following the word and its equivalents, without excluding other elements or objects. The terms “connected,” “linked,” or “connected,” and similar terms are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect.

[0019] It should be noted that the terms "up", "down", "left", "right", "front", and "back" used in this invention are only used to indicate relative positional relationships. When the absolute position of the object being described changes, the relative positional relationship may also change accordingly.

[0020] like Figure 1 The flowchart shown is a method for safe take-off and landing control of unmanned aerial vehicles (UAVs) under strong wind conditions provided in this application embodiment. The method includes:

[0021] Before takeoff, a pre-assessment of the wind field environment and the drone's own status is conducted to obtain a comprehensive takeoff assessment value for the drone. This determines whether the drone's takeoff conditions are met. If they are met, the drone is started with the initial power setting, and the power of the main motor is adjusted according to the takeoff assessment value.

[0022] Preferably, the process for determining whether the conditions for drone takeoff are met is as follows:

[0023] The takeoff evaluation value of the UAV is obtained by coupling the wind shear value of the UAV, the performance evaluation parameters of the UAV, and the wind field sensing value during the pre-inspection period.

[0024] In this embodiment, the wind shear value refers to the change in wind direction angle during the UAV pre-detection period, which quantifies the degree of wind direction change and is measured by the wind direction sensor.

[0025] The drone's takeoff assessment value is compared with the drone's takeoff safety threshold. If the drone's takeoff assessment value is less than the drone's takeoff safety threshold, the drone returns a feedback signal to the control terminal indicating that the takeoff conditions are not met, and waits for the takeoff conditions to be met before taking off.

[0026] If the drone's takeoff assessment value is greater than or equal to the drone's takeoff safety threshold, then the current wind field environment and the drone's own state are deemed to meet the drone's takeoff conditions, and the drone is launched according to the initial power setting.

[0027] The motor power allocation ratio is obtained by mapping the takeoff evaluation value. The specific mapping process is as follows: the takeoff evaluation value is input into a Gaussian function, the calculated base ratio is extracted, and the motor power allocation ratio is output. The power allocation ratio of each motor is coupled with the preset power of the motor. Specifically, the power allocation ratio of each motor is multiplied by the preset power of the motor to complete the adjustment of the main motor power.

[0028] In this embodiment, the motor power allocation ratio is obtained by mapping the takeoff evaluation value. When the takeoff evaluation value is greater than or equal to the takeoff safety threshold of the UAV, the motor is started. The larger the takeoff evaluation value, the better the takeoff conditions of the UAV. The UAV evenly distributes the power of each motor, and the motor power allocation ratio is closer to the minimum value of 0, which conforms to the right half of the Gaussian function distribution curve greater than μ. The closer the takeoff evaluation value is to the takeoff safety threshold, the worse the takeoff conditions of the UAV. The UAV needs to concentrate the power to the main motor used to resist wind interference, and the motor power allocation ratio is closer to the maximum value of 1, which conforms to the characteristic that the Gaussian function distribution curve reaches its maximum value at μ. Therefore, the takeoff evaluation value and the motor power allocation ratio can be fitted using the Gaussian function distribution curve, and a mapping relationship can be constructed by the Gaussian function.

[0029] Preferably, the process of jointly coupling to obtain the takeoff evaluation value of the UAV is as follows:

[0030] Extract the output datasets of each UAV sensor during the pre-inspection period and denote them as the original environmental dataset, including the temperature, humidity, air pressure, and wind speed of the wind field.

[0031] In this embodiment, each of the above-mentioned drone sensors includes a temperature sensor for detecting the ambient temperature of the drone wind field, a humidity sensor for detecting the ambient humidity of the drone wind field, an air pressure sensor for detecting the ambient air pressure of the drone wind field, and a wind speed sensor for detecting the wind speed of the drone wind field.

[0032] The temperature and humidity of the wind field are mapped to a single-valued function using the Magnus formula, and coupled to the actual water vapor pressure. The specific humidity of the wind field is calculated using the specific humidity calculation formula, and the specific humidity of the wind field is coupled to the temperature of the wind field a second time to obtain the virtual temperature of the wind field.

[0033] Based on the virtual temperature of the wind field, the actual air density obtained through the ideal gas law is functionally related to the wind speed of the wind field through the wind resistance equivalent formula, and finally mapped to obtain the equivalent wind speed of the wind field, which is denoted as the wind field induction value.

[0034] It needs to be explained that the above mapping process is based on the virtual temperature of the wind field as input. First, the actual air density is derived by reverse derivation based on the ideal gas law and the actual pressure conditions of the wind field. Then, the equivalent wind resistance formula, which reflects the quantitative relationship between air density, wind speed and wind resistance, is introduced. This formula shows that the core relationship is that wind resistance is positively correlated with air density and the square of wind speed. The actual air density obtained above is substituted into the formula, and the wind speed variable is solved through mathematical operations. Finally, the wind speed value that can equivalently reflect the wind resistance effect of the current wind field is obtained, that is, the equivalent wind speed of the wind field.

[0035] The performance evaluation parameters of the aforementioned drone, including battery capacity, battery voltage, and average power consumption of the drone, are obtained by reading its own parameters through the drone's flight controller.

[0036] The drone's flight time is determined based on its performance evaluation parameters.

[0037] It should be explained that the drone's flight time is obtained by dividing the total energy of the drone's battery by the drone's average power consumption. The total energy of the drone's battery can be directly obtained from the battery capacity and battery voltage. The calculation of the flight time is completed by the flight controller.

[0038] Based on the wind field sensing value, wind shear value, and drone endurance during the drone pre-inspection period, the parameters of each dimension are integrated into a single value. The influencing parameters and normalization results are then weighted and aggregated sequentially to obtain the drone takeoff evaluation value.

[0039] The specific analysis process is as follows:

[0040] ;

[0041] In the formula, TO is the takeoff assessment value of the UAV, and V i Let be the wind field sensing value at the i-th time point during the UAV pre-detection period. , n is the total number of time points in the pre-inspection period, WS is the wind shear value of the UAV pre-inspection period, ld is the UAV's endurance, q1 is the influence parameter corresponding to the wind field sensing value of the UAV pre-inspection period preset in the UAV take-off and landing database, q2 is the characteristic influence parameter corresponding to the wind shear value of the UAV pre-inspection period preset in the UAV take-off and landing database, and q3 is the characteristic influence parameter corresponding to the UAV's endurance preset in the UAV take-off and landing database.

[0042] In this embodiment of the invention, a multivariate analysis of the wind field sensing value, wind direction shear value, and drone endurance during the drone pre-detection period is conducted. Specifically, the correlation between these parameters and the drone flight evaluation value is considered. The wind field sensing value reflects the equivalent wind speed of the wind field, while the wind direction shear value refers to the change in wind direction angle during the drone pre-detection period. Both directly affect the wind field environment in which the drone is located. The wind field sensing value is negatively correlated with the drone's endurance. The larger the wind field sensing value, the greater the actual wind speed that the drone needs to resist, and the shorter the drone's endurance. The wind direction shear value is also negatively correlated with the drone's endurance. The larger the wind direction shear value, the greater the wind direction change in the wind field in which the drone is located, the more power the drone needs to consume to make attitude adjustments, and the shorter the drone's endurance.

[0043] Real-time wind field environmental data is input into the wind field prediction model, and the wind field prediction value is output. The wind field prediction value is calibrated according to the real-time wind field status, and the flight parameters of the UAV are pre-adjusted.

[0044] In this embodiment, the specific construction process of the wind field prediction model is as follows: Real-time collection of wind field environmental data for the pre-detection period; sorting of the wind field environmental data by timestamp; aligning real-time wind speed and real-time wind direction with the timestamps; supplementing the sequence with a relatively uniform sampling frequency through linear interpolation to form a continuous time-series dataset, denoted as the historical wind field environmental dataset. Features strongly correlated with wind field changes are selected, and physical derived features are constructed using the LSTM algorithm to obtain the core features: wind speed change rate and wind direction change amplitude. The correlation between each feature and wind field changes is obtained through the Pearson correlation coefficient. Features with an absolute correlation coefficient greater than a preset correlation threshold are retained, typically those with an absolute value greater than 0.5. The Min-Max method is used to compress all features to a unit interval, converting the one-dimensional time-series data into a three-dimensional tensor required by the LSTM. A custom loss function is defined, including the mean square error of wind speed and wind direction, as well as corresponding physical constraint penalty terms. The Adam optimizer is selected, with mean absolute error as the evaluation metric. EarlyStopping and ModelCheckpoint callback functions are used to achieve early stopping and optimal model saving, resulting in a wind field prediction model. Real-time wind field environmental data is input, and the wind field prediction model outputs the wind field estimate at the next adjustment time point.

[0045] Preferably, the process of calibrating the wind field estimate based on the real-time wind field status is as follows:

[0046] Real-time collection of wind field environmental data, including wind speed and wind direction data.

[0047] The wind field environmental data is sorted by timestamp, and the real-time wind speed and real-time wind direction are aligned with the timestamp. The real-time wind field environmental data during the drone flight is input into the wind field prediction model to obtain the output wind field prediction value.

[0048] The real-time wind field sensing value of the UAV is calculated, and the wind field calibration factor is obtained by mapping the real-time wind field sensing value. The specific mapping process is as follows: the historical wind field prediction value is mapped one-to-one with the real-time wind field sensing value of the historical data, and the ratio of the mean of the fixed deviation of the prediction to the real-time wind field sensing value of the historical data is calculated. The mean of the fixed deviation of the prediction refers to the mean of the difference between the historical wind field prediction value and the real-time wind field sensing value of the historical data, which is denoted as the deviation coefficient. A linear mapping relationship between the real-time wind field sensing value and the fixed deviation value is constructed. The real-time wind field sensing value is input into the linear mapping relationship, and the real-time prediction deviation value is output, which is denoted as the wind field calibration factor.

[0049] In this embodiment, the wind field calibration factor is the degree of difference between the wind field prediction value obtained from historical forecasts and the actual real-time wind field sensing value. It reflects the degree of error that needs to be corrected between the real-time wind field sensing value and the wind field prediction value. The wind field calibration factor is the adjustment parameter for calibrating the wind field prediction value to be closer to the actual wind field environment.

[0050] The wind field calibration factor is coupled with the wind field prediction value by multiplying the wind field calibration factor with the wind field prediction value to obtain the wind field prediction calibration value.

[0051] Based on the wind field prediction calibration value, the wind speed prediction value and the wind direction prediction value of the wind field are obtained.

[0052] Store real-time wind field environmental data during drone flight, update the wind field prediction model at preset intervals, and optimize the model using the newly stored real-time wind field environmental data.

[0053] Preferably, the process of pre-adjusting the various flight parameters of the UAV is as follows:

[0054] The aforementioned flight parameters of the UAV include the motor power and attitude angles of the UAV, which include the yaw angle, pitch angle and roll angle.

[0055] The wind speed estimate of the wind field is coupled with the corresponding preset motor power adjustment coefficient. Specifically, the wind speed estimate of the wind field is multiplied by the preset motor power adjustment coefficient to obtain the estimated motor power of the UAV.

[0056] In this embodiment, the aforementioned preset motor power adjustment coefficient refers to the average value of the ratio of the drone's wind resistance speed to the drone's motor power obtained from historical drone data, reflecting the roughly estimated motor power of the drone under the corresponding wind speed.

[0057] Extract the estimated wind direction angle corresponding to the wind direction estimate of the wind field, compare the difference between the UAV yaw angle and the estimated wind direction angle, and obtain the estimated offset of the UAV yaw angle.

[0058] It should be explained that the estimated wind direction angle corresponding to the above wind direction estimate refers to the wind direction estimate of the wind field expressed as an angle value. The corresponding estimated wind direction angle is the wind direction angle calculated by rotating clockwise from geographical due north.

[0059] The estimated wind resistance of the UAV is calculated based on the wind speed estimate of the wind field. The estimated wind resistance of the UAV is then coupled with the corresponding UAV pitch angle adjustment coefficient. Specifically, the estimated wind resistance of the UAV is multiplied by the corresponding UAV pitch angle adjustment coefficient to obtain the estimated offset of the UAV pitch angle.

[0060] In this embodiment, the aforementioned pitch angle adjustment coefficient of the corresponding UAV refers to the average value of the ratio of the estimated wind resistance experienced by the UAV obtained from historical data of the UAV to the pitch angle adjustment angle of the UAV at this time, reflecting the pitch angle adjustment of the corresponding UAV roughly estimated under the corresponding wind resistance.

[0061] By comprehensively processing the estimated offset of the UAV's yaw angle and the estimated wind speed, the estimated lateral wind speed of the UAV is obtained, and the estimated lateral wind resistance of the UAV is calculated. This is then coupled with the corresponding UAV roll angle adjustment coefficient. Specifically, the estimated lateral wind resistance of the UAV is multiplied by the corresponding UAV roll angle adjustment coefficient to obtain the estimated offset of the UAV roll angle.

[0062] It should be explained that the above-mentioned comprehensive processing of the estimated offset of the drone's yaw angle and the estimated wind speed refers to multiplying the estimated wind speed by the trigonometric function value of the estimated offset of the drone's yaw angle to obtain the estimated lateral wind speed of the drone. The estimated lateral wind resistance of the drone obtained from the above calculation is obtained through the wind resistance calculation formula.

[0063] The estimated motor power of the drone and the estimated offset of each attitude angle of the drone are saved to the memory of the flight controller, and the flight parameters of the drone are adjusted at a constant speed before the next adjustment time point.

[0064] In this embodiment, the flight controller is the core component of the UAV, which can analyze and process the data collected by various sensors to calculate the adjustment amount required for each device. In addition, the flight controller has memory for storing data.

[0065] During flight, the system checks in real time whether the drone has triggered the drone ascent protection mechanism. If it is triggered, the system corrects the pre-adjusted drone flight parameters, dynamically adjusts the takeoff and hovering altitude, and allocates the motor power ratio in real time until the drone flies smoothly to the takeoff and hovering altitude.

[0066] Preferably, the process of triggering the drone ascent protection mechanism is as follows:

[0067] During the process of the UAV flying horizontally to the takeoff hovering altitude, the actual offset of each attitude angle of the UAV is compared with the maximum physical threshold of each attitude angle offset.

[0068] In this embodiment, the aforementioned takeoff hovering altitude refers to the transitional position set during the takeoff phase to determine the subsequent flight status. At this position, the UAV confirms its own status, the surrounding environment, the flight mission, and plans the specific flight trajectory.

[0069] When the attitude angle deviation of any drone is greater than or equal to the maximum physical threshold of each attitude angle, the drone ascent protection mechanism is triggered:

[0070] The relative difference between the actual offset of each attitude angle of the UAV and the maximum physical threshold of each attitude angle offset is calculated to obtain the yaw angle saturation, pitch angle saturation and roll angle saturation.

[0071] It should be explained that when the drone's ascent protection mechanism is triggered, regardless of the relationship between the drone's attitude angle offset and the maximum physical threshold of each attitude angle, the relative difference between the actual offset of each attitude angle and the maximum physical threshold of each attitude angle offset is calculated. When the drone's attitude angle offset is greater than the maximum physical threshold of that attitude angle, the saturation value is positive; when the drone's attitude angle offset is less than the maximum physical threshold of that attitude angle, the saturation value is negative; and when the drone's attitude angle offset is equal to the maximum physical threshold of that attitude angle, the saturation value is zero.

[0072] The saturation values ​​of yaw angle, pitch angle, and roll angle are sorted, and the saturation value with the first value is extracted and recorded as the maximum saturation value of the UAV. The other two saturation values ​​are recorded as the general saturation values ​​of the UAV. The saturation values ​​are integrated to obtain a comprehensive value, which is recorded as the control quantity saturation of the UAV.

[0073] The specific analysis process is as follows:

[0074] ;

[0075] In the formula, S is the control saturation of the UAV, SM is the maximum saturation of the UAV, a is the risk amplification coefficient of the UAV control, b is the risk triggering threshold of the UAV, c is the collaborative risk coefficient of the UAV control, S1 is the general saturation of one UAV, and S2 is the general saturation of the other UAV.

[0076] The saturation of the UAV's control input is input into the memory of the flight controller and coupled with the estimated offset of each attitude angle of the UAV to obtain the reasonable offset of each attitude angle of the UAV, and update the average adjustment of the UAV.

[0077] In this embodiment, the control saturation of the UAV is coupled with the estimated offset of each attitude angle of the UAV in order to comprehensively adjust the control of each attitude angle of the UAV. Since a certain attitude angle of the UAV exceeds the maximum physical threshold of the UAV, it is necessary to adjust each attitude angle in a balanced manner to prevent the change of wind field environment from causing the UAV's flight state to oscillate when the saturated attitude angle is significantly corrected.

[0078] Based on the reasonable offset of the UAV's various attitude angles, the power distribution of each flight control motor is adjusted accordingly:

[0079] Increase the power of each flight control motor corresponding to the attitude angle with positive saturation, and couple the corresponding estimated motor power with the control quantity saturation of the UAV. Specifically, multiply the corresponding estimated motor power with the control quantity saturation of the UAV to obtain the increased motor power value.

[0080] The power of each flight control motor corresponding to the attitude angle with negative saturation is reduced. The corresponding estimated motor power is coupled with the control quantity saturation of the UAV. Specifically, the corresponding estimated motor power is multiplied by the control quantity saturation of the UAV to obtain the motor power reduction value.

[0081] The power of each flight control motor remains unchanged at attitude angles where the saturation is zero.

[0082] Preferably, the mechanism for triggering the drone's ascent protection also includes:

[0083] The duration corresponding to a certain attitude angle offset being greater than or equal to the maximum physical threshold of the attitude angle is recorded as the saturation duration of the UAV control quantity.

[0084] Based on the saturation duration of each UAV control quantity, the gust frequency of the wind field within the protection mechanism execution cycle is obtained.

[0085] It needs to be explained that in a gusty environment, the drone is affected by the constantly changing wind speed and direction, and the drone's control quantity will be in a saturated state. Therefore, the saturation time of the drone can be equivalent to the duration of the gusts in the wind field, thus obtaining the gust frequency of the wind field within the protection mechanism's execution cycle.

[0086] The hovering altitude adjustment factor is obtained by mapping the gust frequency of the wind field. The specific mapping process is as follows: The correlation coefficient between the captured gust frequency and the drone's altitude interference is calculated to verify the correlation. The normal equation is used to calculate the minimum sum of squared residuals between the gust frequency and the drone's altitude interference through matrix operations, directly obtaining a unique coefficient vector. The reliability of this mapping relationship is verified by evaluating the R-squared value. The gust frequency of the wind field is then substituted into the mapping to output the optimal coefficient, obtaining the hovering altitude adjustment factor. This hovering altitude adjustment factor is then coupled with a preset takeoff hovering altitude by multiplying the hovering altitude by the preset takeoff hovering altitude to obtain the takeoff hovering altitude adjustment value.

[0087] The maximum takeoff hovering height is obtained by superimposing the preset takeoff hovering height with the takeoff hovering height adjustment value. The minimum takeoff hovering height is obtained by subtracting the takeoff hovering height from the takeoff hovering height adjustment value. The takeoff hovering range of the UAV is obtained from the maximum takeoff hovering height and the minimum takeoff hovering height. When the UAV reaches the takeoff hovering range, the takeoff process of the UAV is considered to be completed.

[0088] Before the drone lands, a pre-assessment is performed to obtain the drone's landing assessment value. Based on the landing assessment value, it is determined whether to land. If landing is selected, the near-ground wind speed is collected when the drone reaches the deceleration point to determine the wind field type. The landing mode is adaptively matched according to the wind field type to enable the drone to land smoothly and complete the drone take-off and landing control task.

[0089] Preferably, the adaptive matching landing mode process based on wind field type is as follows:

[0090] The system acquires a preset identification standard for drones that detects continuous or gusty winds, and configures a stability range for near-ground wind speeds, including a first stable wind speed range, a second stable wind speed range, and a third stable wind speed range, which correspond to the type of wind field the drone is in during its descent: continuous wind, gusty wind, and mixed wind, respectively.

[0091] It should be explained that the first stable wind speed range corresponds to continuous wind, which is a high wind speed range; the second stable wind speed range corresponds to mixed wind, which is a wind speed fluctuation range; and the third stable wind speed range corresponds to gusts, which is a low wind speed range.

[0092] In this embodiment, the preset identification criteria for continuous wind or gusts of wind for drones are determined by the specific configuration of the drone, and the preset identification criteria for continuous wind or gusts of wind are based on historical data of the actual flight process of the drone.

[0093] When the drone's sensors detect the drone's landing at a preset altitude near-ground deceleration point, they acquire real-time near-ground wind speed data to determine the type of wind field in which the drone is located.

[0094] When the wind field where the drone is located is a continuous wind, the ground effect adaptive control mode is activated:

[0095] The ground effect curve stored in the UAV take-off and landing control system is extracted. The ground effect curve is the ground effect interference value obtained based on the historical ground effect influence results of the UAV. It represents the motor power consumed by the UAV to resist the ground effect when the UAV lands at different near-ground altitudes.

[0096] According to the embodiments of the present invention Figure 2 The curve in the figure indicates the motor power consumed by the drone to counteract the ground effect as it descends to a corresponding altitude. Figure 2 This is the ground effect curve stored in the UAV take-off and landing control system provided in this embodiment of the invention.

[0097] When the drone descends to the corresponding altitude, the flight controller releases the pre-stored ground effect interference value and adjusts the drone motor speed uniformly according to the ground effect interference value.

[0098] The downward-looking sensor outputs the current height change rate in real time. The difference between the current height change rate and the reference change rate is compared, and the relative difference between the current height change rate and the reference change rate is calculated and recorded as the ground effect difference.

[0099] It should be explained that the real-time output of the current altitude change rate by the downward-looking sensor refers to extracting the dataset of the drone's altitude above the ground collected by the downward-looking sensor, calculating the altitude divided by time, obtaining the current altitude change rate, and then outputting it in real time.

[0100] The ground effect difference is compared with the preset ground effect warning value. If the ground effect difference is greater than or equal to the ground effect warning value, it is determined that the UAV's flight altitude has changed rapidly, and the ground effect warning is activated. The reference thrust is obtained based on the total weight of the UAV. The ground effect difference is extracted and coupled with the reference thrust. Specifically, the ground effect difference is multiplied by the reference thrust to calculate the thrust error of the UAV.

[0101] The thrust error of the drone is equivalent to wind resistance, and the correction value of the drone motor speed is calculated to adjust the motor power of the drone.

[0102] It should be explained that equating the thrust error of a drone with wind resistance means using the wind resistance calculation formula to convert the thrust of the drone caused by the ground effect into the motor power that the drone needs to consume to resist the thrust.

[0103] When the wind field where the drone is located is characterized by gusts and mixed winds, activate the S-shaped micro-maneuver mode.

[0104] The preferred S-shaped micro-maneuvering mode has the following specific control process:

[0105] Calculate the proportion of each wind direction appearing during the drone's descent to the total effective wind direction data, and sort them in descending order of the proportion to obtain the first sequence of wind direction proportions.

[0106] The duration of each wind direction is counted, and the duration of each wind direction is coupled with the corresponding ordinal number of the first sequence of wind direction proportions to obtain the second ordinal number of each wind direction proportion. The second ordinal numbers of each wind direction proportion are sorted in descending order to obtain the wind direction proportion sequence. The wind direction that ranks first in the wind direction proportion sequence is extracted and recorded as the main wind direction.

[0107] Based on the main direction of the wind field, the change in peak gust is extracted and recorded as the UAV disturbance value.

[0108] In this embodiment, the aforementioned gust peak change refers to the change in the maximum wind speed in the wind field where the UAV is located, based on the main direction, and obtained through the UAV's wind speed sensor.

[0109] Based on the disturbance value of the UAV, an S-shaped maneuver is performed to resist the wind field. At the same time, the attitude deviation of the UAV is collected. During the resistance process, the descent protection mechanism is automatically triggered to ensure the attitude safety of the UAV.

[0110] It should be explained that the descent protection mechanism is similar to the ascent protection mechanism. The triggering condition is also to compare the actual offset of each attitude angle of the UAV with the maximum physical threshold of each attitude angle offset. When the offset of any UAV attitude angle is greater than or equal to the maximum physical threshold of each attitude angle, the protection mechanism is triggered. The content of the descent protection mechanism is also the same as that of the ascent protection mechanism, which is a protection mechanism for the UAV during the landing process.

[0111] Preferably, completing the drone takeoff and landing control task also includes:

[0112] During the take-off and landing of the drone, the drone records the change in peak gust in real time and compares the change in peak gust with the preset peak change safety threshold.

[0113] When the change in peak gust is greater than or equal to the safe threshold for peak gust change, a sudden change in wind force is detected, and a strong wind change signal is immediately sent to the flight control system. At the same time, the UAV triggers a cooperative response procedure:

[0114] The system outputs the power change rate and motor temperature in real time. Based on the intensity level of sudden strong winds, it adjusts the preset safety thresholds for the drone's power and temperature, and records the adjusted safety thresholds for the drone's power and temperature as the actual safety thresholds.

[0115] It should be explained that the aforementioned adjustment of the drone's battery and temperature preset safety thresholds based on the intensity level of the sudden strong wind means that the intensity level of the sudden strong wind is matched with the preset safe wind force level, which is divided into a high intensity level greater than the safe wind force and a safe intensity level. These correspond to the safety thresholds for the drone's battery and temperature for the high intensity level and the safe intensity level for the safe drone, respectively. If the intensity level of the sudden strong wind is less than or equal to the preset safe wind force level, the safety thresholds for the drone's battery and temperature remain equal to the preset safety thresholds. If the intensity level of the sudden strong wind is greater than the preset safe wind force level, the safety thresholds for the drone's battery and temperature are adjusted to the safety thresholds for the high intensity level.

[0116] If the rate of change of output power and the motor temperature are greater than or equal to the actual safety threshold, the take-off and landing process of the drone will be immediately suspended. If the drone is in the take-off process, it will return to the starting point. If the drone is in the landing process, it will go back to flight.

[0117] The drone's takeoff and landing operations will only resume once the drone's output power change rate and motor temperature have decreased to below the actual safety threshold.

[0118] The following points need to be explained:

[0119] (1) The accompanying drawings of the embodiments of the present invention only involve the structures involved in the embodiments of the present invention. Other structures can refer to the general design.

[0120] (2) For clarity, the thickness of layers or regions is enlarged or reduced in the drawings used to describe embodiments of the invention, i.e., these drawings are not drawn to scale. It is understood that when an element such as a layer, film, region or substrate is referred to as being “above” or “below” another element, the element may be “directly” located “above” or “below” the other element or there may be intermediate elements.

[0121] (3) Where there is no conflict, the embodiments of the present invention and the features in the embodiments can be combined with each other to obtain new embodiments.

[0122] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. The scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for safe take-off and landing control of unmanned aerial vehicles (UAVs) under strong wind conditions, characterized in that, include: Before takeoff, a pre-assessment of the wind field environment and the drone's own status is conducted to obtain a comprehensive takeoff assessment value for the drone. If the conditions for takeoff are met, the drone is started according to the initial power setting, and the power of the main motor is adjusted according to the takeoff assessment value. Real-time wind field environmental data is input into the wind field prediction model, and the wind field prediction value is output. The wind field prediction value is calibrated according to the real-time wind field status, and the flight parameters of the UAV are pre-adjusted. During flight, the drone is checked in real time to see if it has triggered the drone ascent protection mechanism. If it is triggered, the pre-adjusted drone flight parameters are corrected, and the take-off and hovering altitude is dynamically adjusted. The motor power ratio is allocated in real time until the drone flies smoothly to the take-off and hovering altitude. The specific process for triggering the drone's ascent protection mechanism is as follows: During the process of the UAV flying horizontally to the takeoff hovering altitude, the actual offset of each attitude angle of the UAV is compared with the maximum physical threshold of each attitude angle offset. When the attitude angle deviation of any drone is greater than or equal to the maximum physical threshold of each attitude angle, the drone ascent protection mechanism is triggered: The relative difference between the actual offset of each attitude angle of the UAV and the maximum physical threshold of each attitude angle offset is calculated to obtain the yaw angle saturation, pitch angle saturation and roll angle saturation. The saturation values ​​of yaw angle, pitch angle, and roll angle are sorted, and the saturation value with the first sort is extracted and recorded as the maximum saturation value of the UAV. The other two saturation values ​​are recorded as the general saturation values ​​of the UAV. The saturation values ​​are integrated to obtain a comprehensive value, which is recorded as the control quantity saturation of the UAV. The control saturation of the UAV is input into the control unit of the flight controller and coupled with the estimated offset of each attitude angle of the UAV to obtain the reasonable offset of each attitude angle of the UAV and update the average adjustment of the UAV. Based on the reasonable offset of the UAV's various attitude angles, the power distribution of each flight control motor is adjusted accordingly: Increase the power of each flight control motor corresponding to the attitude angle with positive saturation, and couple the corresponding estimated motor power with the saturation of the UAV's control quantity to obtain the motor power increase value; The power of each flight control motor corresponding to the attitude angle with negative saturation is reduced, and the corresponding estimated motor power is coupled with the saturation of the UAV's control quantity to obtain the motor power reduction value. The power of each flight control motor remains unchanged for attitude angles where the saturation is zero. The mechanism for triggering the drone ascent protection also includes: The duration corresponding to a certain attitude angle offset being greater than or equal to the maximum physical threshold of the attitude angle is recorded as the saturation duration of the UAV control quantity. Based on the saturation duration of each UAV control quantity, the gust frequency of the wind field within the protection mechanism execution cycle is obtained; The hovering altitude adjustment factor is obtained by mapping the gust frequency of the wind field. The hovering altitude adjustment factor is coupled with the preset takeoff hovering altitude to obtain the takeoff hovering altitude adjustment value. The preset takeoff hovering height is superimposed with the takeoff hovering height adjustment value to obtain the maximum takeoff hovering height threshold. The takeoff hovering height is subtracted from the takeoff hovering height adjustment value to obtain the minimum takeoff hovering height threshold. The takeoff hovering range of the UAV is obtained from the maximum takeoff hovering height threshold and the minimum takeoff hovering height threshold. When the UAV reaches the takeoff hovering range, the UAV takeoff process is considered to be complete. Before the drone lands, a pre-assessment is performed to obtain the drone's landing assessment value. Based on the landing assessment value, it is determined whether to land. If landing is selected, the near-ground wind speed is collected when the drone reaches the deceleration point to determine the wind field type. The landing mode is adaptively matched according to the wind field type to enable the drone to land smoothly and complete the drone take-off and landing control task.

2. The method for safe take-off and landing control of unmanned aerial vehicles under strong wind conditions according to claim 1, characterized in that, The specific process for determining whether the conditions for drone takeoff are met is as follows: The takeoff evaluation value of the UAV is obtained by coupling the wind shear value of the UAV, the performance evaluation parameters of the UAV and the wind field sensing value during the pre-inspection period. The drone's takeoff assessment value is compared with the drone's takeoff safety threshold. If the drone's takeoff assessment value is less than the drone's takeoff safety threshold, the drone returns a feedback signal to the control terminal indicating that the takeoff conditions are not met, and waits for the takeoff conditions to be met before taking off. If the drone's takeoff assessment value is greater than or equal to the drone's takeoff safety threshold, then the current wind field environment and the drone's own state are deemed to meet the drone's takeoff conditions, and the drone is started according to the initial power setting. The motor power allocation ratio is obtained by mapping the takeoff evaluation value. The power allocation ratio of each motor is coupled with the preset power of the motor to complete the adjustment of the main motor power.

3. The method for safe take-off and landing control of unmanned aerial vehicles under strong wind conditions according to claim 2, characterized in that, The takeoff evaluation value of the UAV obtained through the joint coupling is calculated as follows: Extract the output datasets of each UAV sensor during the pre-inspection period and denote them as the original environmental dataset, including the temperature, humidity, air pressure, and wind speed of the wind field. The temperature and humidity of the wind field are mapped to a single-valued function using the Magnus formula, and coupled to the actual water vapor pressure. The specific humidity of the wind field is calculated using the specific humidity calculation formula, and the specific humidity of the wind field is coupled to the temperature of the wind field a second time to obtain the virtual temperature of the wind field. Based on the virtual temperature of the wind field, the actual air density obtained through the ideal gas law is functionally related to the wind speed of the wind field through the wind resistance equivalent formula, and finally mapped to obtain the equivalent wind speed of the wind field, which is denoted as the wind field induction value. The performance evaluation parameters of the drone include battery capacity, battery voltage, and average power consumption of the drone. The drone's flight time is obtained based on its performance evaluation parameters. Based on the wind field sensing value, wind shear value, and drone endurance during the drone pre-inspection period, the parameters of each dimension are integrated into a single value. The influencing parameters and normalization results are then weighted and aggregated sequentially to obtain the drone takeoff evaluation value.

4. The method for safe take-off and landing control of unmanned aerial vehicles under strong wind conditions according to claim 1, characterized in that, The specific analysis process for calibrating the wind field prediction based on the real-time wind field status is as follows: Real-time acquisition of wind field environmental data, including wind speed and wind direction data; The wind field environmental data is sorted by timestamp, and the real-time wind speed and real-time wind direction are aligned with the timestamp. The real-time wind field environmental data during the drone flight is input into the wind field prediction model to obtain the output wind field prediction value. Calculate the real-time wind field sensing value of the UAV, and obtain the wind field calibration factor based on the real-time wind field sensing value; The wind field calibration factor is coupled with the wind field prediction value to obtain the wind field prediction calibration value; Based on the wind field prediction calibration value, the wind speed prediction value and the wind direction prediction value of the wind field can be obtained. Store real-time wind field environmental data during drone flight, update the wind field prediction model at preset intervals, and optimize the model using the newly stored real-time wind field environmental data.

5. The method for safe take-off and landing control of unmanned aerial vehicles under strong wind conditions according to claim 1, characterized in that, The specific adjustment process for the pre-adjustment of various flight parameters of the UAV is as follows: The flight parameters of the UAV include the motor power of the UAV and the attitude angles of the UAV, which include the yaw angle, pitch angle and roll angle of the UAV. The wind speed estimate of the wind field is coupled with the corresponding preset motor power adjustment coefficient to obtain the estimated motor power of the UAV. Extract the estimated wind direction angle corresponding to the wind direction estimate of the wind field, compare the difference between the UAV yaw angle and the estimated wind direction angle, and obtain the estimated offset of the UAV yaw angle. The estimated wind resistance of the UAV is calculated based on the wind speed estimate of the wind field. The estimated wind resistance of the UAV is then coupled with the corresponding UAV pitch angle adjustment coefficient to obtain the estimated offset of the UAV pitch angle. By comprehensively processing the estimated offset of the UAV's yaw angle and the estimated wind speed, the estimated lateral wind speed of the UAV is obtained, the estimated lateral wind resistance of the UAV is calculated, and coupled with the corresponding UAV roll angle adjustment coefficient, the estimated offset of the UAV's roll angle is obtained. The estimated motor power of the drone and the estimated offset of each attitude angle of the drone are saved to the memory of the flight controller, and the flight parameters of the drone are adjusted at a constant speed before the next adjustment time point.

6. The method for safe take-off and landing control of unmanned aerial vehicles under strong wind conditions according to claim 1, characterized in that, The adaptive matching of landing modes based on wind field type is specifically performed as follows: The drone's preset identification criteria for continuous wind or gust wind are obtained, and the stability range of near-ground wind speed is configured, including the first wind speed stability range, the second wind speed stability range and the third wind speed stability range, which correspond to the type of wind field where the drone is located during the landing process, namely continuous wind, gust wind and mixed wind, respectively. When the drone's sensors detect the drone's landing at a preset altitude near-ground deceleration point, they acquire real-time near-ground wind speed data to determine the type of wind field in which the drone is located. When the wind field where the drone is located is a continuous wind, the ground effect adaptive control mode is activated: Extract the ground effect influence curve stored in the UAV take-off and landing control system. The ground effect influence curve is the ground effect interference value obtained based on the historical ground effect influence results of the UAV, which represents the motor power consumed by the UAV to resist the ground effect when the UAV lands at different near-ground altitudes. When the drone descends to the corresponding altitude, the flight controller releases the pre-stored ground effect interference value and adjusts the drone motor speed uniformly according to the ground effect interference value. The downward-looking sensor outputs the current height change rate in real time. The difference between the current height change rate and the reference change rate is compared, and the relative difference between the current height change rate and the reference change rate is calculated and recorded as the ground effect difference. The ground effect difference is compared with the preset ground effect warning value. If the ground effect difference is greater than or equal to the ground effect warning value, it is determined that the UAV's flight altitude has changed rapidly, and the ground effect warning is activated. The reference thrust is obtained based on the total weight of the UAV. The ground effect difference is extracted and coupled with the reference thrust to calculate the thrust error of the UAV. The thrust error of the drone is equivalent to wind resistance, and the correction value of the drone motor speed is calculated to adjust the motor power of the drone. When the wind field where the drone is located is characterized by gusts and mixed winds, activate the S-shaped micro-maneuver mode.

7. The method for safe take-off and landing control of unmanned aerial vehicles under strong wind conditions according to claim 6, characterized in that, The specific control process of the S-shaped micro-maneuvering mode is as follows: Calculate the proportion of the number of times each wind direction appears during the drone's landing process to the total effective wind direction data, and sort them in descending order of the proportion to obtain the first sequence of wind direction proportions; The duration of each wind direction is counted. The duration of each wind direction is coupled with the corresponding ordinal number of the first sequence of wind direction proportions to obtain the second ordinal number of each wind direction proportion. The second ordinal numbers of each wind direction proportion are sorted in descending order to obtain the wind direction proportion sequence. The wind direction that ranks first in the wind direction proportion sequence is extracted and recorded as the main wind field direction. Based on the main direction of the wind field, the change in peak gust is extracted and recorded as the UAV disturbance value; Based on the disturbance value of the UAV, an S-shaped maneuver is performed to resist the wind field. At the same time, the attitude deviation of the UAV is collected. During the resistance process, the descent protection mechanism is automatically triggered to ensure the attitude safety of the UAV.

8. The method for safe take-off and landing control of unmanned aerial vehicles under strong wind conditions according to claim 1, characterized in that, The completion of the UAV take-off and landing control task also includes: During the take-off and landing of the drone, the drone records the change in peak gust in real time and compares the change in peak gust with the preset peak change safety threshold. When the change in peak gust is greater than or equal to the safe threshold for peak gust change, a sudden change in wind force is detected, and a strong wind change signal is immediately sent to the flight control system. At the same time, the UAV triggers a cooperative response procedure: The system outputs the rate of change in battery power and motor temperature in real time. Based on the intensity level of sudden strong winds, it adjusts the preset safety thresholds for the drone's battery power and temperature, and records the adjusted safety thresholds for the drone's battery power and temperature as the actual safety thresholds. If the rate of change of output power and the motor temperature are greater than or equal to the actual safety threshold, the take-off and landing process of the drone will be immediately suspended. If the drone is in the take-off process, it will return to the starting point. If the drone is in the landing process, it will go back to flight. The drone's takeoff and landing operations will only resume once the drone's output power change rate and motor temperature have decreased to below the actual safety threshold.

Images