A method and system for early warning of risks of hoisting unmanned aerial vehicles

Abstract

This invention relates to the field of control, specifically to a method and system for early warning of lifting risks in unmanned aerial vehicles (UAVs). The method includes: calculating the load balance index of the UAV; executing a primary warning when the load balance index is less than a preset balance threshold; marking the UAV with the smallest load as an abnormal UAV and the other UAVs as normal UAVs; when executing the primary warning, constructing a time series and calculating the angular velocity change coefficient of the abnormal UAV at each moment; calculating the difference between the relative positions of the abnormal UAV and the normal UAV at two adjacent moments, and summing all differences to obtain the relative movement value of the abnormal UAV at the corresponding moment; calculating the rollover warning index of the abnormal UAV, and executing a secondary warning when the rollover warning index is greater than a preset rollover threshold. This invention effectively monitors abnormal situations of UAVs, improving the safety and efficiency of lifting operations.

Description

Technical Field

[0001] The present invention relates to the field of control, and more specifically, to a method and system for early warning of risks of hoisting a drone. Background Art

[0002] A lifting drone is a specially designed drone that has a strong load-bearing capacity and stable flight performance, and can perform lifting tasks in various complex environments. This type of drone usually adopts a multi-rotor configuration, has multiple motor drives, and can hover and adjust in the air to adapt to changes in wind and airflow. Their main features include strong load-bearing capacity, good stability, high flexibility, high efficiency, easy maintenance and high safety.

[0003] At present, the existing technology, such as the patent application document with the publication number CN113657665A, discloses a UAV system status monitoring and early warning system based on artificial intelligence, including: data collection module, data analysis and cleaning module, system database, artificial intelligence neural network and early warning response module; the data collection module is used to collect the status of the UAV itself and the surrounding environment data; the data analysis and cleaning module analyzes and processes the real-time data collected by the data collection module and transfers it to the system database for storage, and continuously expands and updates the database to form big data, realizes the intelligent update of data, and increases the amount of data collection; when the system database saves the real-time flight data, it inputs the data into the artificially constructed deep neural network for deep learning to complete the training of the deep neural network; the early warning response module selects whether to start the early warning response according to the analysis results obtained by the comparison in the artificial intelligence neural network. The UAV system status monitoring and early warning system monitors the status of the UAV operating at high altitude in real time and provides early warning and intervention for the possible risks of the UAV.

[0004] However, within a certain range, there are drones working at the same time, and the system disclosed in the above patent application document cannot distinguish between normal drones and abnormal drones, resulting in reduced accuracy of early warning, and further affecting the safety and efficiency of drone operations. Summary of the invention

[0005] In order to solve the above-mentioned technical problem of reduced warning accuracy, which in turn leads to reduced safety and efficiency of UAV operations, the present invention provides solutions in the following aspects.

[0006] In a first aspect, a method for early warning of a risk of hoisting a drone comprises:

[0007] Obtain the loads of multiple hoisting drones and calculate the load balance index; when the load balance index is less than the preset balance threshold, execute the primary warning; and mark the hoisting drone with the smallest load as an abnormal drone, and mark other hoisting drones as normal drones;

[0008] When executing the primary warning, the angular velocity of the abnormal UAV is monitored, a time series is constructed, and the angular velocity change coefficient of the abnormal UAV at each moment in the time series is calculated;

[0009] Obtain the relative position of the abnormal UAV and all other UAVs at each moment in the time series, calculate the difference between the relative positions of the abnormal UAV and the normal UAV at two adjacent moments, and sum up all the differences to obtain the relative movement value of the abnormal UAV at the corresponding moment;

[0010] The rollover warning index of the abnormal UAV is calculated, and when the rollover warning index is greater than a preset rollover threshold, a secondary warning is executed; the rollover warning index is positively correlated with the relative movement value and the angular velocity change coefficient.

[0011] The present invention firstly performs a primary warning by analyzing the load distribution of multiple lifting drones. On the basis of the primary warning, the angular velocity change of the abnormal drone and the position change of the abnormal drone relative to other drones are monitored, and then a rollover warning index of the abnormal drone is calculated. A secondary warning is made according to the rollover warning index, thereby effectively monitoring and warning abnormal situations that may occur during the collaborative lifting process of multiple lifting drones, thereby improving the safety and efficiency of the lifting operation.

[0012] Preferably, the performing of the second-level warning further includes:

[0013] Reduce the operating speed and flight altitude of abnormal drones.

[0014] Preferably, the process of obtaining the load balancing index includes:

[0015] Calculate the ratio of the minimum load to the maximum load of all the hoisting drones at the current moment as the first ratio;

[0016] Calculate the average value of the sum of the ratios of the minimum load to the maximum load of all the hoisting drones at a set number of moments before the current moment as the second ratio;

[0017] The difference between the first ratio and the second ratio is used as the load balancing index.

[0018] The first ratio reflects the load distribution at the current moment, while the second ratio reflects the average load distribution over a period of time. By calculating the difference between these two ratios, it is possible to detect whether there has been a sudden change in the load distribution, which may be due to a drone malfunction or other abnormal situation.

[0019] Preferably, the process of obtaining the angular velocity variation coefficient includes:

[0020] Obtain the angular velocity vector of the abnormal UAV and the angular velocity vector of the normal UAV at any moment in the time series;

[0021] Calculating the cosine of the angle between the two angular velocity vectors;

[0022] Calculate the ratio of the modulus of the angular velocity vector of the abnormal UAV to the modulus of the angular velocity vector of the normal UAV;

[0023] The maximum value of the product of the cosine value and the ratio is used as the angular velocity variation coefficient.

[0024] By calculating the angular velocity change coefficient of the abnormal drone and comparing it with the angular velocity characteristics of the normal drone, the abnormal state of the drone can be identified. This comparison can reflect the significant change in the angular velocity of the drone, indicating potential mechanical failure, sensor problem or control system abnormality.

[0025] Preferably, the process of acquiring the rollover warning index includes:

[0026] Acquire the sequence of the angular velocity variation coefficients and the sequence of the relative movement values ​​in the window according to the length of the preset time window in the time series;

[0027] The Pearson correlation coefficient between the sequence of the angular velocity variation coefficients and the sequence of the relative movement values ​​is used as the rollover warning index.

[0028] The angular velocity change coefficient reflects how fast the abnormal UAV's posture changes within a certain period of time, while the relative movement value represents the position change of the abnormal UAV relative to a certain reference state. Combining these two parameters can more comprehensively evaluate the current working status of the abnormal UAV.

[0029] Traditional rollover warning systems may rely on a single warning indicator, such as roll angle or lateral acceleration. However, these indicators may not be accurate enough in certain situations, resulting in false positives or negatives. By combining multiple parameters and calculating the Pearson correlation coefficient, the rollover warning index can reduce this error and improve the overall performance of the system.

[0030] Preferably, the process of acquiring the rollover warning index further includes:

[0031] Calculate the ratio of the relative movement values ​​corresponding to the start time and the end time in the window;

[0032] The Pearson correlation coefficient is multiplied by the ratio of the relative movement values ​​corresponding to the start time and the end time in the window to obtain an adjusted Pearson correlation coefficient, and the adjusted Pearson correlation coefficient is used as the rollover warning index.

[0033] By combining the Pearson correlation coefficient with the ratio of the relative movement value, a more accurate and reliable rollover warning index can be obtained. This adjusted index can better reflect the dynamic changes of abnormal drones during flight, thereby improving the accuracy and timeliness of warnings.

[0034] Preferably, the process of obtaining the load balancing index includes:

[0035] Calculate the variance of all the hoisting drone loads at the current moment, recorded as the first variance;

[0036] Calculate the average value of the variance of the loads in all the hoisting drones at a set number of moments before the current moment, and record it as the second variance;

[0037] The difference between the first variance and the second variance is used as the load balancing index.

[0038] In a second aspect, a lifting risk warning system for lifting unmanned aerial vehicles includes: a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the above-mentioned lifting risk warning method for lifting unmanned aerial vehicles is implemented.

[0039] The beneficial effects of the present invention are:

[0040] The present invention first monitors the load distribution of the lifting UAV. In the case of unbalanced load and primary warning, the system will mark the UAV with the smallest load as an abnormal UAV, further monitor the abnormal UAV, and calculate the rollover warning index based on the relative movement value and angular velocity change coefficient of the abnormal UAV, and then warn of possible risks, thereby improving the safety and reliability of the lifting operation. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] By reading the following detailed description with reference to the accompanying drawings, the above and other objects, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-limiting manner, and the same or corresponding reference numerals represent the same or corresponding parts, wherein:

[0042] Figure 1It is a method flow chart of steps S1 to S4 in a method for early warning of a risk of hoisting a drone according to an embodiment of the present invention. DETAILED DESCRIPTION

[0043] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative work are within the scope of protection of the present invention.

[0044] The embodiment of the present invention discloses a method for early warning of the risk of hoisting a drone, referring to Figure 1 , including steps S1 to S4, which are specifically as follows:

[0045] S1: Obtain the loads of multiple lifting drones and calculate the load balance index; when the load balance index is less than the preset balance threshold, execute the primary warning; and mark the lifting drone with the smallest load as an abnormal drone, and mark the other lifting drones as normal drones.

[0046] When multiple lifting drones work together, the center of gravity of the load may deflect due to changes in the external environment (such as wind) or the cargo lifting and bundling. This deflection will cause the force on the drone to change, resulting in asynchronous loads between multiple drones, that is, some drones have too large a load capacity and may exceed their maximum load limit; while some drones have too small a load and are easily affected by external factors such as wind, further causing asynchronous drone operations.

[0047] In the embodiment of the present invention, the number of hoisting drones is set to K, and the value of K can be 4, 5 or 6.

[0048] Therefore, the load distribution of the hoisting drone is analyzed. The general steps of the analysis are as follows:

[0049] Calculate the ratio of the minimum load to the maximum load of all the hoisting drones at the current moment as the first ratio;

[0050] Calculate the average value of the sum of the ratios of the minimum load to the maximum load of all the hoisting drones at a set number of moments before the current moment as the second ratio;

[0051] The difference between the first ratio and the second ratio is used as the load balance index of the hoisting drone.

[0052] Then the load balance index of the above-mentioned hoisting UAV satisfies the relationship:

[0053]

[0054] In the formula, is the load balance index of the hoisting drone, For the The first crane load of the moment, For the Moment The load of the crane drone, For the The first crane load of the moment, For the Moment The load of the crane drone, Based on The set number of moments before the moment. Among them, is the minimum value, is the maximum value.

[0055] The load balance index is determined by the load balance at a single moment and the average of the load balance at multiple moments. If the load at a certain moment is very unbalanced (that is, the ratio of the minimum to the maximum value is very small), or the average load balance at multiple moments is very poor, then the value of the load balance index will be small. On the contrary, if the load at all moments is very balanced, then the value of the load balance index will be close to 1.

[0056] In another embodiment, another method for calculating the load balance index of a hoisting drone is provided.

[0057] Calculate the variance of all the hoisting drone loads at the current moment, recorded as the first variance;

[0058] Calculate the average value of the variance of the loads in all the hoisting drones at a set number of moments before the current moment, and record it as the second variance;

[0059] The difference between the first variance and the second variance is used as the load balancing index to satisfy the relationship:

[0060]

[0061] In the formula, is the load balance index of the hoisting drone, For the Moment The load of the crane drone, For the Moment The load of the crane drone, Based on A set number of moments before the moment, For the The average load of all hoisting drones at the moment, No. The average load of all lifting drones at the moment.

[0062] in, is the first variance, is the second variance.

[0063] The time unit is seconds.

[0064] If you need to monitor and evaluate the changing trend of the load balancing status for a long time, the first method of calculating the load balancing index may be more appropriate. If you need to quickly determine the balance of the current load distribution and it is not sensitive to time changes, the second method of calculating the load balancing index may be more appropriate.

[0065] After calculating the load balance index of the hoisting drone, a balance threshold is set. When the load balance index is less than the balance threshold, a primary warning is executed.

[0066] The balance threshold is 0.7. It should be noted that the balance threshold can be adjusted dynamically based on the test results and actual operation experience. If it is found that the threshold of 0.7 is too loose or too strict, it can be adjusted appropriately according to the actual situation to ensure the safety and stability of the flight.

[0067] When a hoisting drone is performing a mission, the balance of the load is directly related to the stability and safety of the flight. If the load is unevenly distributed, it may cause the drone to become unstable during flight, or even cause a safety accident. Therefore, by setting a balance threshold and a primary warning mechanism, the problem of load imbalance can be discovered and handled in a timely manner, thereby ensuring the flight safety of the drone.

[0068] In addition, the lifting drone with the smallest load is marked as an abnormal drone, and the other lifting drones are marked as normal drones. During the lifting operation, the failure or abnormality of any drone may threaten the safety and stability of the entire operation. By marking the drone with the smallest load as abnormal and taking timely measures to deal with it, this risk can be minimized. At the same time, this also reminds the operator to remain vigilant throughout the operation and pay close attention to the status of each drone to ensure the safety and stability of the overall operation.

[0069] S2: When executing the primary warning, the angular velocity of the abnormal UAV is monitored, a time series is constructed, and the angular velocity change coefficient of the abnormal UAV at each moment in the time series is calculated.

[0070] When a UAV has an abnormal load, its angular velocity may change, which may reflect some mechanical or control system problems of the UAV. By calculating the angular velocity change coefficient, this change can be quantified, thereby more accurately diagnosing the abnormal state of the UAV.

[0071] When the primary warning is performed in the above step S1, the angular velocity of the abnormal UAV needs to be monitored in real time, which can be achieved by an inertial measurement unit (IMU) such as a gyroscope installed on the UAV, which can sensitively measure the angular velocity of the UAV.

[0072] The monitored angular velocity data are arranged in chronological order to form a time series. This time series will record the angular velocity changes of the drone at different time points and provide basic data for subsequent analysis. The real-time monitoring time is set to 1000 seconds.

[0073] First, for any abnormal drone at any moment in the time series, obtain its angular velocity vector At the same time, obtain the angular velocity vector of the normal drone at the same time .

[0074] Then, use the vector dot product formula to calculate the angle between the two angular velocity vectors The cosine of :

[0075]

[0076] in, is the dot product of two vectors, and are the magnitudes of the two vectors respectively.

[0077] Furthermore, the ratio of the modulus of the angular velocity vector of the abnormal UAV to the modulus of the angular velocity vector of the normal UAV is calculated.

[0078] Furthermore, the cosine value obtained above is multiplied by the ratio of the modulus to obtain a coefficient.

[0079] According to the above calculation process, a coefficient corresponding to the abnormal UAV and all normal UAVs can be obtained in the same way, and then the maximum value of all coefficients is used as the angular velocity change coefficient of the abnormal UAV at the corresponding moment, and then the angular velocity change coefficient of the abnormal UAV at all moments in the time series is obtained.

[0080] In another embodiment, instead of comparing the direction and magnitude differences of the angular velocity vectors between an abnormal UAV and a normal UAV, a method is provided for analyzing the dynamic changes of the angular velocity of the abnormal UAV to calculate the angular velocity change coefficient of the abnormal UAV.

[0081] Then the angular velocity change coefficient of the abnormal UAV satisfies the relationship:

[0082]

[0083] In the formula, For the abnormal drone The coefficient of change of angular velocity at time, For the abnormal drone The angular velocity at time, For the abnormal drone Angular velocity at time.

[0084] Here, when calculating the angular velocity change coefficient, the angular velocity changes of the three axes are combined into a scalar value.

[0085] By calculating the cosine value of the angle between the two angular velocity vectors, the angular difference between the angular velocity vectors of the abnormal drone and the normal drone can be quantified. A cosine value close to 1 means the angle is close to 0 degrees, that is, the directions of the angular velocity vectors are similar; a cosine value close to 0 means the angle is close to 90 degrees, that is, the directions of the angular velocity vectors are perpendicular; a cosine value close to -1 means the angle is close to 180 degrees, that is, the directions of the angular velocity vectors are opposite.

[0086] By calculating the ratio of the magnitude of the angular velocity vector of the abnormal drone to the magnitude of the angular velocity vector of the normal drone, the difference in magnitude between the two can be quantified. This ratio can help understand whether the angular velocity of the abnormal drone is within the normal range, or whether there is an abnormality that is too large or too small.

[0087] By comparing the coefficients corresponding to the abnormal UAV with those of all normal UAVs and taking the maximum value as the angular velocity change coefficient of the abnormal UAV at the corresponding moment, the most significant abnormal situation can be identified.

[0088] S3: Obtain the relative position of the abnormal UAV and all other UAVs at each moment in the time series, calculate the difference between the relative positions of the abnormal UAV and the normal UAVs at two adjacent moments, and sum up all the differences to obtain the relative movement value of the abnormal UAV at the corresponding moment.

[0089] In order to evaluate the anti-wind resistance of the drone under a specific load, it is necessary to monitor the change of the angular velocity coefficient. If the angular velocity coefficient increases rapidly and continuously, it may indicate that the drone encounters strong wind resistance or other disturbances, which is highly dangerous.

[0090] An infrared transmitter and a receiver are arranged on each hoisting UAV. In the time sequence of the above step S2, for each moment, the position of the abnormal UAV and the positions of all other normal UAVs are firstly obtained.

[0091] For each pair of abnormal drones and normal drones, the relative position difference between them at two adjacent moments (e.g., the current moment and the previous moment) is calculated. This difference can be a difference on a two-dimensional plane (if the drone moves on a plane) or a difference in three-dimensional space (if the drone moves in the air).

[0092] For each pair of relative position differences between the abnormal UAV and all normal UAVs, they are added together to obtain the relative movement value of the abnormal UAV at that moment.

[0093] Based on the relative movement value of the abnormal drone at that moment calculated above, the relative movement values ​​corresponding to the abnormal drones at all moments in the time series can be calculated similarly.

[0094] The magnitude of the relative movement value can be used as an indicator to assess the risk level of abnormal drones. If the relative movement value suddenly increases, it may indicate that the drone has encountered an emergency or operational error, and immediate action is required.

[0095] S4: Calculate the rollover warning index of the abnormal UAV. When the rollover warning index is greater than a preset rollover threshold, execute a secondary warning. The rollover warning index is positively correlated with the relative movement value and the angular velocity change coefficient.

[0096] In time series analysis, we first extract data within a specific window from the data stream based on the preset time window length. This window can be a fixed time period, such as the past 5 seconds, 10 seconds, etc.

[0097] In this time window, two sequences are obtained: the sequence of angular velocity change coefficients and the sequence of relative movement values. The sequence of angular velocity change coefficients records the change of the angular velocity of the drone at each moment in the window, while the sequence of relative movement values ​​records the change of the relative position of the drone at each moment in the window.

[0098] The Pearson correlation coefficient is used to measure the correlation between the angular velocity change coefficient sequence and the relative movement value sequence. If the correlation coefficient is high, it means that there is a strong correlation between the angular velocity change and the relative movement value, which may indicate that the abnormal drone has a risk of rollover. Therefore, the Pearson correlation coefficient between the calculated angular velocity change coefficient sequence and the relative movement value sequence is used as the rollover warning index of the abnormal drone.

[0099] Furthermore, a factor is introduced to adjust the calculated Pearson correlation coefficient. The steps for obtaining the factor are as follows:

[0100] Calculate the ratio of the relative movement values ​​corresponding to the start time and the end time in the above window;

[0101] The Pearson correlation coefficient is multiplied by the ratio of the relative movement values ​​corresponding to the start time and the end time in the window to obtain an adjusted Pearson correlation coefficient, and the adjusted Pearson correlation coefficient is used as the rollover warning index. This index combines the correlation between the angular velocity change and the relative movement and the amplitude of the relative movement, so that the rollover risk can be more comprehensively evaluated.

[0102] When the rollover warning index is greater than the preset rollover threshold (the rollover threshold can be determined by analyzing historical flight data to determine the range of changes in the rollover index of the drone in normal flight and rollover conditions, which can be used as a reference for setting the threshold), a secondary warning is executed, and the operator can take measures to reduce the flight speed of the abnormal drone, and can also reduce potential risks by lowering the flight altitude of the abnormal drone.

[0103] An embodiment of the present invention also discloses a lifting risk warning system for lifting a drone, including a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a lifting risk warning method for lifting a drone according to the present invention is implemented.

[0104] The system also includes other components familiar to those skilled in the art, such as a communication bus and a communication interface. The configuration and functions of these components are known in the art and will not be described in detail here.

[0105] In the present invention, the aforementioned memory may be any tangible medium containing or storing a program, which may be used by or in combination with an instruction execution system, apparatus or device. For example, a computer-readable storage medium may be any appropriate magnetic storage medium or magneto-optical storage medium, such as a resistive random access memory RRAM (Resistive Random Access Memory), a dynamic random access memory DRAM (Dynamic Random Access Memory), a static random access memory SRAM (Static Random-Access Memory), an enhanced dynamic random access memory EDRAM (Enhanced Dynamic Random Access Memory), a high-bandwidth memory HBM (High-Bandwidth Memory), a hybrid memory cube HMC (Hybrid Memory Cube), etc., or any other medium that can be used to store the required information and can be accessed by an application, a module or both. Any such computer storage medium may be part of a device or accessible or connectable to a device.

[0106] In the description of this specification, "plurality" or "several" means at least two, such as two, three or more, etc., unless otherwise clearly and specifically defined.

[0107] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, various alternatives to the embodiments of the present invention described herein may be employed.

Claims

1. A method for early warning of the risk of hoisting a drone, characterized in that: include: Obtain the loads of multiple hoisting drones and calculate the load balance index; When the load balance index is less than the preset balance threshold, a primary warning is executed; The lifting drone with the smallest load is marked as an abnormal drone, and the other lifting drones are marked as normal drones; When executing the primary warning, the angular velocity of the abnormal UAV is monitored, a time series is constructed, and the angular velocity change coefficient of the abnormal UAV at each moment in the time series is calculated; Obtain the relative position of the abnormal UAV and all other UAVs at each moment in the time series, calculate the difference between the relative positions of the abnormal UAV and the normal UAV at two adjacent moments, and sum up all the differences to obtain the relative movement value of the abnormal UAV at the corresponding moment; Calculate the rollover warning index of the abnormal drone. When the rollover warning index is greater than the preset rollover threshold, execute the second-level warning. The rollover warning index is positively correlated with the relative movement value and the angular velocity variation coefficient.

2. A method for early warning of risks in hoisting a drone according to claim 1, characterized in that: The implementation of the second-level warning also includes: Reduce the operating speed and flight altitude of abnormal drones.

3. A method for early warning of risk of hoisting a drone according to claim 2, characterized in that: The process of obtaining the load balancing index includes: Calculate the ratio of the minimum load to the maximum load of all the hoisting drones at the current moment as the first ratio; Calculate the average value of the sum of the ratios of the minimum load to the maximum load of all the hoisting drones at a set number of moments before the current moment as the second ratio; The difference between the first ratio and the second ratio is used as the load balancing index.

4. The method for early warning of risk of hoisting a drone according to claim 3, characterized in that: The process of obtaining the angular velocity variation coefficient includes: Obtain the angular velocity vector of the abnormal UAV and the angular velocity vector of the normal UAV at any moment in the time series; Calculating the cosine of the angle between the two angular velocity vectors; Calculate the ratio of the modulus of the angular velocity vector of the abnormal UAV to the modulus of the angular velocity vector of the normal UAV; The maximum value of the product of the cosine value and the ratio is used as the angular velocity variation coefficient.

5. The method for early warning of risk of hoisting a drone according to claim 4, characterized in that: The process of obtaining the rollover warning index includes: Acquire the sequence of the angular velocity variation coefficients and the sequence of the relative movement values ​​in the window according to the length of the preset time window in the time series; The Pearson correlation coefficient between the sequence of the angular velocity variation coefficients and the sequence of the relative movement values ​​is used as the rollover warning index.

6. A method for early warning of risk of hoisting a drone according to claim 5, characterized in that: The process of obtaining the rollover warning index also includes: Calculate the ratio of the relative movement values ​​corresponding to the start time and the end time in the window; The Pearson correlation coefficient is multiplied by the ratio of the relative movement values ​​corresponding to the start time and the end time in the window to obtain an adjusted Pearson correlation coefficient, and the adjusted Pearson correlation coefficient is used as the rollover warning index.

7. The method for early warning of risk of hoisting a drone according to claim 2, characterized in that: The process of obtaining the load balancing index includes: Calculate the variance of all the hoisting drone loads at the current moment, recorded as the first variance; Calculate the average value of the variance of the loads in all the hoisting drones at a set number of moments before the current moment, and record it as the second variance; The difference between the first variance and the second variance is used as the load balancing index.

8. A risk warning system for hoisting drones, characterized in that: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the method for early warning of the risk of hoisting a drone according to any one of claims 1 to 7 is implemented.

Images