Data processing method and system based on airspace monitoring of unmanned aerial vehicle

Abstract

The invention discloses a data processing method and system based on unmanned aerial vehicle airspace monitoring. The method comprises the following steps: acquiring real-time motion parameter data and timestamps of each unmanned aerial vehicle and aircraft in an unmanned aerial vehicle group; according to the motion parameter data and the timestamp, carrying out data point equal-interval sampling operation to obtain a time sequence data set; performing data cleaning operation according to the time sequence data set to obtain screened data; according to the screening data, performing normalization operation to obtain normalized data; performing relative collision risk calculation according to the normalized data to obtain relative collision risk data; according to the relative collision risk data and the time sequence data set, performing path planning matrix construction operation to obtain risk target point set data; and performing conflict-free screening operation according to the risk target point set data to obtain a target point. The method can reduce the collision risk of the unmanned aerial vehicle.

Description

Technical Field

[0001] The present invention relates to the field of unmanned aerial vehicle control technology, and in particular to a data processing method and system based on unmanned aerial vehicle airspace monitoring. Background Art

[0002] At present, a drone refers to an unmanned aerial vehicle controlled by a radio remote control device or its own program control device. It is a type of aircraft. The difference between a drone and a spacecraft is that a spacecraft usually flies at high altitudes and outside the atmosphere, while a drone flies close to the ground and is often used in military, entertainment, surveying and mapping and other fields. The flight airspace of a drone refers to the altitude at which the drone is located during its flight activities according to the flight plan and the spatial range of the flight mission. There are still many collision risks when existing drones fly, especially with the continuous expansion of civil drone applications, unsafe incidents between drones and other aircraft such as manned aircraft are gradually increasing, and the risk of collision between the two is becoming more and more serious. In order to improve the safety of drone flights, it is particularly important to monitor the drone's airspace and process and analyze the data.

[0003] In one existing technology, historical collision data is divided into training data and verification data, and trained through a neural network model to obtain a prediction classification model, and the flight collision risk of civil drones is predicted based on data characteristics. Existing application scenarios have high requirements for real-time collision risk detection of drones, but the existing technology does not involve the detection and analysis of real-time data of drones, and therefore cannot effectively reduce the collision risk of drones during real-time flight. Summary of the invention

[0004] The present invention provides a data processing method and system based on unmanned aerial vehicle airspace monitoring, so as to reduce the collision risk of unmanned aerial vehicles during flight.

[0005] In the first aspect, in order to solve the above technical problems, the present invention provides a data processing method based on drone airspace monitoring, comprising:

[0006] Obtain real-time motion parameter data and timestamps for each drone and aircraft in the drone swarm;

[0007] According to the motion parameter data and the timestamp, a data point sampling operation with equal spacing is performed to obtain a time series data set;

[0008] According to the time series data set, a data cleaning operation is performed to obtain screening data;

[0009] Performing a normalization operation according to the screened data to obtain normalized data;

[0010] performing relative collision risk calculation according to the normalized data to obtain relative collision risk data;

[0011] Performing a path planning matrix construction operation according to the relative collision risk data and the time series data set to obtain risk target point set data;

[0012] According to the risk target point set data, a conflict-free screening operation is performed to obtain the target point.

[0013] In an optional implementation, performing an equidistant sampling operation of data points according to the motion parameter data and the timestamp to obtain a time series data set includes:

[0014] The motion parameter data includes position, velocity and acceleration;

[0015] The time series dataset is constructed as follows:

[0016] X={X0,X1,X2,…,X t ,…,X T}

[0017] Among them, X t is a multidimensional dataset at time t; T represents the number of collected time series;

[0018] Each X t It can be further expressed as:

[0019] X t =[u1(t) u2(t) … u M (t) a1(t) a2(t) … a N (t)] T

[0020] Among them, u i (t) is the motion parameter data of the i-th UAV at time t, a j (t) is the motion parameter data of the jth aircraft at time t, M is the total number of drones, and N is the total number of aircraft.

[0021] In an optional implementation, performing a data cleaning operation according to the time series data set to obtain screening data includes:

[0022] According to the time series data set, a data segment selection operation is performed to obtain time segment data;

[0023] Performing a clustering operation according to the time segment data to obtain time clustering data;

[0024] According to the time clustering data, a Laida detection operation is performed to obtain screening data.

[0025] In an optional implementation, performing a Laida detection operation according to the time clustering data to obtain screening data includes:

[0026] Performing a residual error calculation operation according to the time clustering data to obtain a residual error;

[0027] Performing a standard deviation calculation operation according to the time clustering data to obtain a standard deviation;

[0028] According to the residual error and the standard deviation, when the absolute value of the residual error is greater than three times the standard deviation, a deletion operation is performed to obtain screening data;

[0029] The residual error calculation formula is as follows:

[0030]

[0031] Among them, v i represents the residual error of the i-th data; x i Represents the i-th data value; represents the arithmetic mean;

[0032] The standard deviation calculation formula is as follows:

[0033]

[0034] Where σ represents the standard deviation; x i Represents the i-th data value; represents the arithmetic mean; n represents the total number of data.

[0035] In an optional implementation, performing a normalization operation on the screening data to obtain normalized data includes:

[0036] Perform feature selection based on the screening data to obtain similar feature data;

[0037] Performing a feature normalization operation according to the similar feature data to obtain normalized data;

[0038] The calculation formula of the feature normalization operation is as follows:

[0039] X nnnn,k =norm(X k )

[0040] Among them, X k is the similar feature data of the drone group at time k, norm(·) is the normalization function, X nnnn,k Represents the normalized data of the drone swarm at time k.

[0041] In an optional implementation, performing relative collision risk calculation according to the normalized data to obtain relative collision risk data includes:

[0042] Calculating the distance between the UAV and the aircraft based on the normalized data to obtain distance data;

[0043] performing relative collision risk calculation according to the normalized data and the distance data to obtain relative collision risk data;

[0044] The relative collision risk calculation formula is as follows:

[0045]

[0046] Among them, R ik represents the relative collision risk value between the i-th UAV and the k-th aircraft; d ik represents the distance value between the i-th UAV and the k-th aircraft; |u i | represents the magnitude of the acceleration vector of the i-th drone; |a k | represents the magnitude of the kth aircraft acceleration vector.

[0047] In an optional implementation, a path planning matrix construction operation is performed according to the relative collision risk data and the time series data set to obtain risk target point set data, including:

[0048] Extracting the drone location information based on the time series data set to obtain the drone real-time location data;

[0049] According to the real-time position data of the UAV, path planning is performed to obtain target point set data;

[0050] Performing a risk information embedding operation according to the target point set data and the relative collision risk data to obtain risk target point set data;

[0051] The risk target point set data is as follows:

[0052] P i ={p1,p2,…p j ,…,p m}

[0053] Among them, P i represents the i-th drone risk target point set; p j represents the location data of the jth target point embedded with risk information; m represents the number of elements in the risk target point set.

[0054] In an optional implementation, performing a conflict-free screening operation according to the risk target point set data to obtain the target point includes:

[0055] Performing a comprehensive information evaluation based on the risk target point set data to obtain a comprehensive information evaluation value;

[0056] Sorting the risk target point set data from small to large based on all the comprehensive information evaluation values ​​to obtain ordered target point set data;

[0057] According to the ordered target point set data, sequentially selecting ordered target points as initial target points;

[0058] A radius threshold is set. When it is determined that there are other drones within the radius of the initial target point, the initial target point is determined to be a drone collision point and the initial target point is discarded; when it is determined that there are no other drones within the radius of the initial target point, the initial target point is determined to be the target point;

[0059] The calculation formula for comprehensive information evaluation is as follows:

[0060]

[0061] Among them, V i represents the comprehensive information evaluation value of the i-th UAV; N represents the total number of aircraft; R ik represents the relative collision risk value between the i-th UAV and the k-th aircraft; d ik Represents the distance value between the i-th UAV and the k-th aircraft.

[0062] In a second aspect, the present invention provides a data processing system based on drone airspace monitoring, comprising:

[0063] An input module is used to obtain real-time motion parameter data and timestamps of each drone and aircraft in the drone swarm;

[0064] A sampling module, used for performing an equidistant sampling operation of data points according to the motion parameter data and the timestamp to obtain a time series data set;

[0065] A screening module, used for performing a data cleaning operation according to the time series data set to obtain screening data;

[0066] A normalization module, used to perform a normalization operation based on the screening data to obtain normalized data;

[0067] A risk assessment module, used to calculate the relative collision risk according to the normalized data to obtain relative collision risk data;

[0068] A target point construction module, used to perform a path planning matrix construction operation according to the relative collision risk data and the time series data set to obtain risk target point set data;

[0069] The output module is used to perform a conflict-free screening operation based on the risk target point set data to obtain the target point.

[0070] In a third aspect, the present invention also provides an electronic device, comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, it implements any one of the above-mentioned data processing methods based on drone airspace monitoring.

[0071] Compared with the prior art, the present invention has the following beneficial effects:

[0072] The present invention provides a data processing method based on drone airspace monitoring, characterized in that it includes: obtaining real-time motion parameter data and timestamps of each drone and aircraft in a drone group; performing an equidistant sampling operation of data points according to the motion parameter data and the timestamp to obtain a time series data set; performing a data cleaning operation according to the time series data set to obtain screening data; performing a normalization operation according to the screening data to obtain normalized data; performing a relative collision risk calculation according to the normalized data to obtain relative collision risk data; performing a path planning matrix construction operation according to the relative collision risk data and the time series data set to obtain risk target point set data; performing a conflict-free screening operation according to the risk target point set data to obtain a target point. The present invention can reduce the risk of drone flight collision by evaluating the real-time risk during the drone flight, embedding the risk information into the path points, and finally performing a conflict-free screening operation to obtain the target points. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] Figure 1 It is a schematic flow chart of a method for data processing based on drone airspace monitoring provided by an embodiment of the present invention;

[0074] Figure 2 It is a schematic diagram of the system structure of data processing based on drone airspace monitoring provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0075] 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 only 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 ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0076] Reference Figure 1The embodiment of the present invention provides a data processing method based on drone airspace monitoring, comprising the following steps:

[0077] S11, obtaining real-time motion parameter data and timestamps of each drone and aircraft in the drone group;

[0078] S12, performing an equal-interval sampling operation on data points according to the motion parameter data and the timestamp to obtain a time series data set;

[0079] S13, performing a data cleaning operation according to the time series data set to obtain screening data;

[0080] S14, performing a normalization operation according to the screened data to obtain normalized data;

[0081] S15, performing relative collision risk calculation according to the normalized data to obtain relative collision risk data;

[0082] S16, performing a path planning matrix construction operation according to the relative collision risk data and the time series data set to obtain risk target point set data;

[0083] S17, performing a conflict-free screening operation according to the risk target point set data to obtain the target point.

[0084] In step S11, it should be noted that real-time motion parameter data refers to information such as the position, speed and acceleration of drones and aircraft at a specific moment, and these parameters are crucial for understanding and predicting the motion state of these aircraft. Timestamps refer to specific time points corresponding to these motion parameter data, expressed in the form of date and time, and used to identify the exact moment of data recording. The process of acquiring these features involves sensor data acquisition and time synchronization technology. Sensors equipped on drones and aircraft, including GPS, accelerometers and gyroscopes, can monitor and record parameters such as the position and speed of aircraft in real time. The data of these sensors are processed by the internal system of the aircraft and stored or transmitted together with the current timestamp. The acquisition of timestamps depends on the internal clock of the aircraft or the external time synchronization signal, such as GPS time, to ensure the accuracy and consistency of time. The operation of these real-time motion parameter data and timestamps involves real-time data acquisition, processing and storage. The control system of the aircraft will continuously read data from the sensor, and combine the timestamp information to package these data into data frames or data packets, and then send them to the ground control station or data processing center through a wireless communication link. At the data processing center, these data will be further analyzed and processed to monitor the status of the aircraft, predict possible conflicts, and optimize the flight path. In subsequent applications, these real-time motion parameter data with timestamps can be used for a variety of operations, including path planning and conflict avoidance of drone swarms. These data can be used to predict future positions and speeds, thereby calculating potential conflict points and planning conflict-avoiding paths in advance. In addition, these data can also be used for flight performance analysis, flight safety monitoring, and flight log recording and backtracking. Through in-depth analysis of these data, the operating efficiency and safety of drone swarms can be improved.

[0085] In step S12, a data point sampling operation with equal spacing is performed according to the motion parameter data and the timestamp to obtain a time series data set, including:

[0086] The motion parameter data includes position, velocity and acceleration;

[0087] The time series dataset is constructed as follows:

[0088] X={X0,X1,X2,…,X t ,…,X T}

[0089] Among them, X t is a multidimensional dataset at time t; T represents the number of collected time series;

[0090] Each X t It can be further expressed as:

[0091] X t =[u1(t) u2(t) …u i (t) … u M (t) a1(t) a2(t) …a j (t) … a N (t)] T

[0092] Among them, u i (t) is the motion parameter data of the i-th UAV at time t, a j (t) is the motion parameter data of the jth aircraft at time t, M is the total number of drones, and N is the total number of aircraft.

[0093] It should be noted that the motion parameter data in step S12 covers the position, velocity and acceleration of the drone and aircraft. These parameters are key information describing the dynamic characteristics of the aircraft. They are collected in real time by the sensors on the aircraft and recorded and stored together with the corresponding timestamps through the aircraft's data processing system. The timestamp is a specific time point corresponding to the motion parameter data, which is used to identify the collection moment of each data point to ensure the timeliness and traceability of the data. The construction of the time series data set is achieved through the operation of equally spaced data point sampling. This operation involves arranging the collected discrete motion parameter data in chronological order to form a continuous time series. The time series data set X is expressed as X=X0,X1,X2,…,X t ,…,X T , where X t is a multidimensional dataset at time t, where T represents the number of time series collected. This representation method enables the dataset to reflect the dynamic characteristics of changes over time, providing a basis for subsequent data analysis and processing. Each X_t can be further represented as a vector containing the motion parameters of the drone and aircraft, specifically X i =[u1(t),u2(t),…,u M (t),a1(t),a2(t),…,a N (t)] T , where u i (t) is the motion parameter data of the i-th UAV at time t, a j (t) is the motion parameter data of the jth aircraft at time t, M is the total number of drones, and N is the total number of aircraft. This representation method makes the data set X at each time point t tBoth can provide complete motion status information of drone swarms and aircraft. The acquisition and construction of these features are achieved through the collaborative work of the aircraft's sensor system, data acquisition system, and data processing system. The sensor system is responsible for real-time acquisition of the aircraft's position, velocity, and acceleration data, the data acquisition system is responsible for recording and storing these data together with the corresponding timestamps, and the data processing system is responsible for collating and analyzing these data in chronological order to form a time series data set. In subsequent applications, these features and constructed time series data sets can be used for a variety of operations, including aircraft trajectory analysis, motion state prediction, flight safety monitoring, flight path planning, and conflict avoidance. Through in-depth analysis of time series data sets, the operating efficiency and safety of aircraft can be improved, flight plans can be optimized, flight risks can be reduced, and data support can be provided for aircraft maintenance and upgrades.

[0094] In step S13, a data cleaning operation is performed according to the time series data set to obtain screening data, including:

[0095] According to the time series data set, a data segment selection operation is performed to obtain time segment data;

[0096] Performing a clustering operation according to the time segment data to obtain time clustering data;

[0097] According to the time clustering data, a Laida detection operation is performed to obtain screening data.

[0098] Among them, according to the time clustering data, a Laida detection operation is performed to obtain screening data, including:

[0099] Performing a residual error calculation operation according to the time clustering data to obtain a residual error;

[0100] Performing a standard deviation calculation operation according to the time clustering data to obtain a standard deviation;

[0101] According to the residual error and the standard deviation, when the absolute value of the residual error is greater than three times the standard deviation, a deletion operation is performed to obtain screening data;

[0102] The residual error calculation formula is as follows:

[0103]

[0104] Among them, v i represents the residual error of the u-th data; x i Represents the i-th data value; represents the arithmetic mean;

[0105] The standard deviation calculation formula is as follows:

[0106]

[0107] Where σ represents the standard deviation; x i Represents the i-th data value; represents the arithmetic mean; n represents the total number of data.

[0108] It should be noted that in step S13, the time series data set is a set consisting of a timestamp and corresponding motion parameter data, which is obtained from the original data through an equally spaced sampling operation, and contains the position, velocity and acceleration information of the drone and aircraft at continuous time points. The time segment data is a subset obtained by selecting data of a specific time period from the time series data set, which represents the motion state of the aircraft within a specific time window. For example, the time segment selected in the present invention is composed of five adjacent moment data. Of course, time segments of other time lengths can also be selected according to different user needs or different application scenarios, and the present invention is not limited to this. The clustering operation is the process of grouping the time segment data. Through the k-means clustering algorithm, similar data points are classified into one category to form time clustering data. The Laida detection operation is a statistical method based on identifying and processing outliers. In this process, the residual error is first calculated, which represents the deviation of each data point from the arithmetic mean of the data set. The calculation formula is: where v i represents the residual error of the i-th data, x i represents the i-th data value, represents the arithmetic mean. Next, calculate the standard deviation, which is a statistic that measures the degree of dispersion of the data. The calculation formula is Where σ represents the standard deviation, x i represents the i-th data value, The table represents the arithmetic mean, and n represents the total number of data. The standard deviation can quantify the degree of deviation of the data point from the mean. After obtaining the residual error and standard deviation, by comparing the absolute value of the residual error of each data point with the size of three times the standard deviation, it can be determined whether the data point is an outlier. If the absolute value of the residual error of a data point is greater than three times the standard deviation, the data point is considered to be abnormal and deleted, and finally the filtered data is obtained. The filtered data is a clean data set obtained after the data cleaning operation, which excludes outliers and improves the quality and reliability of the data. These filtered data can be used for subsequent data analysis, model training and decision support to ensure the accuracy and effectiveness of the analysis results. In this way, outliers can be effectively identified and removed from the time series data set, providing more accurate and reliable data processing results for drone airspace monitoring.

[0109] In step S14, a normalization operation is performed according to the screened data to obtain normalized data, including:

[0110] Perform feature selection based on the screening data to obtain similar feature data;

[0111] Performing a feature normalization operation according to the similar feature data to obtain normalized data;

[0112] The calculation formula of the feature normalization operation is as follows:

[0113] X nnnn,k =norm(X k )

[0114] Among them, X k is the similar feature data of the drone group at time k, norm(X k ) is the normalization function, X nnnn,k Represents the normalized data of the drone swarm at time k.

[0115] It should be noted that in step S14, the filtered data refers to the data set after data cleaning, in which outliers have been removed, and these data are the basis for subsequent analysis and processing. Similar feature data refers to data with the same attributes or types in the filtered data, for example, the location data of all drones or the speed data of all aircraft. The feature selection operation is the process of extracting similar feature data from the filtered data. The purpose of this step is to simplify the data structure for subsequent processing. The feature normalization operation scales the similar feature data so that it falls within the range of [-1,1] to eliminate the influence of the dimensions between different features and improve the convergence speed and accuracy of the algorithm. The normalization calculation formula is X nnnk =norm(X k ), where X k is the similar feature data of the drone group at time k, norm(·) is the normalization function, X nnnk Represents the normalized data of the drone swarm at time k. The normalization function can use different algorithms. In the example of the present invention, the minimum-maximum normalization method is selected. Of course, other normalization methods can be selected according to different user needs or different application scenarios, such as Z-score normalization, etc., which are not limited by the present invention. The execution of the normalization operation involves applying a normalization function to each similar feature data to convert the original data into data of a uniform scale. This process can be implemented by programming. The normalized data can be used for subsequent statistical analysis because all features have the same scale, and the normalized data can improve the performance and accuracy of these methods. In addition, normalized data also helps to more intuitively compare the relationship between different features when visualizing the data.

[0116] In step S15, relative collision risk calculation is performed based on the normalized data to obtain relative collision risk data, including:

[0117] Calculating the distance between the UAV and the aircraft based on the normalized data to obtain distance data;

[0118] performing relative collision risk calculation according to the normalized data and the distance data to obtain relative collision risk data;

[0119] The relative collision risk calculation formula is as follows:

[0120]

[0121] Among them, R ik represents the relative collision risk value between the i-th UAV and the k-th aircraft; d ik represents the distance value between the i-th UAV and the k-th aircraft; |u i | represents the magnitude of the acceleration vector of the i-th drone; |a k | represents the magnitude of the kth aircraft acceleration vector.

[0122] It should be noted that in step S15, the normalized data is a data set that has been normalized, and includes the position, velocity, and acceleration information of the UAV and the aircraft at each time, and these data have been scaled to a uniform scale. The distance data is obtained by calculating the distance between the UAV and the aircraft, and this calculation is based on the position information in the normalized data. Specifically, the Euclidean distance formula can be used to calculate the straight-line distance between two objects, that is, where x i ,y i ,z i is the position coordinate of the ith UAV, (x k ,y k ,z k ) is the position coordinate of the kth aircraft. The relative collision risk data is calculated based on the normalized data and distance data, which quantifies the possibility of a collision between the drone and the aircraft. The calculation formula for the relative collision risk is Where R k represents the relative collision risk value between the i-th UAV and the k-th aircraft, d k is the distance between the two, |u i | is the modulus of the acceleration vector of the ith drone, |a k | is the modulus of the acceleration vector of the kth aircraft. The modulus of the acceleration vector can be obtained by calculating the Euclidean norm of the acceleration vector, that is, and where u i ,u i ,u i is the acceleration component of the ith UAV, a i ,a i ,a i is the acceleration component of the kth aircraft. These calculation operations are implemented through programming. The corresponding functions can be defined in data processing software or programming languages ​​and applied to the normalized data set. Relative collision risk data can be used as a basis for flight safety monitoring and path planning to help identify potential collision risks and take preventive measures. In practical applications, these data can be used to adjust the flight paths of drones and aircraft to avoid high-risk approaches and ensure flight safety.

[0123] In step S16, a path planning matrix construction operation is performed according to the relative collision risk data and the time series data set to obtain risk target point set data, including:

[0124] Extracting the drone location information based on the time series data set to obtain the drone real-time location data;

[0125] According to the real-time position data of the UAV, path planning is performed to obtain target point set data;

[0126] Performing a risk information embedding operation according to the target point set data and the relative collision risk data to obtain risk target point set data;

[0127] The risk target point set data is as follows:

[0128] P i ={p1,p2,…p j ,…,p m}

[0129] Among them, P i represents the i-th drone risk target point set; p j represents the location data of the jth target point embedded with risk information; m represents the number of elements in the risk target point set.

[0130] It should be noted that in step S16, the relative collision risk data is calculated based on the distance and acceleration vector modulus between the UAV and the aircraft, and quantifies the potential collision risk. The time series data set contains the position, velocity and acceleration information of the UAV and the aircraft at continuous time points. This information is collected in real time by sensors and sampled at equal intervals. The real-time position data of the UAV is extracted from the time series data set, reflecting the spatial position of the UAV at a specific time point. Path planning is based on the real-time position data of the UAV. The purpose is to plan a path from the current position to the target position for the UAV, and obtain target point set data, which includes key position points that the UAV needs to pass during the flight. The risk information embedding operation is the process of combining the relative collision risk data with the target point set data. The purpose is to mark the collision risk level associated with each target point on each target point to form risk target point set data. Risk target point set data P i ={p1,p2,…,p j ,…,p m} represents the risk target point set of the i-th drone, where p j represents the location data of the jth target point embedded with risk information, and m represents the number of elements in the risk target point set. The risk target point set data can be used as a basis for UAV flight path planning and decision support, helping UAV operators and automatic control systems to identify high-risk areas and take corresponding avoidance measures. In practical applications, these data can be used to dynamically adjust the flight path, avoid high-risk areas, and ensure flight safety.

[0131] In step S17, a conflict-free screening operation is performed according to the risk target point set data to obtain target points, including:

[0132] Performing a comprehensive information evaluation based on the risk target point set data to obtain a comprehensive information evaluation value;

[0133] Sorting the risk target point set data from small to large based on all the comprehensive information evaluation values ​​to obtain ordered target point set data;

[0134] According to the ordered target point set data, sequentially selecting ordered target points as initial target points;

[0135] A radius threshold is set. When it is determined that there are other drones within the radius of the initial target point, the initial target point is determined to be a drone collision point and the initial target point is discarded; when it is determined that there are no other drones within the radius of the initial target point, the initial target point is determined to be the target point;

[0136] The calculation formula for comprehensive information evaluation is as follows:

[0137]

[0138] Among them, V i represents the comprehensive information evaluation value of the i-th UAV; N represents the total number of aircraft; R ik represents the relative collision risk value between the i-th UAV and the k-th aircraft; d ik Represents the distance value between the i-th UAV and the k-th aircraft.

[0139] It should be noted that in step S17, the risk target point set data is constructed based on the positions of the drone and the aircraft, the relative collision risk values, and the distance values ​​between them, where each target point is embedded with corresponding risk information. i is the relative collision risk value R between the i-th UAV and all aircraft k and the distance value d k The weighted sum and average are calculated as follows: Where N is the total number of aircraft. This evaluation value comprehensively considers the relative collision risk and distance between the drone and the aircraft, and provides a quantitative risk measurement for each drone. The ordered target point set data is obtained by sorting the risk target point set based on the comprehensive information evaluation value. The basis for sorting can be the size of the comprehensive information evaluation value, thereby obtaining a target point set sorted according to the risk level. This ordered target point set data helps to prioritize those target points with higher risks. UAV conflict detection is based on the ordered target point set data, and the purpose is to identify and exclude those target points that may cause conflicts. The drone collision detection method adopted by the present invention is to set a radius threshold. When it is determined that there are other drones within the radius of the target point, the target point is determined to be a drone collision point and discarded. In the present invention, the threshold radius is selected as 50m. Of course, according to different user needs or different application scenarios, other radius thresholds can also be selected, and the present invention does not limit this. These operations are implemented by algorithms, and corresponding functions can be defined in programming languages ​​and applied to risk target point set data. The result of the conflict-free screening operation, that is, the target point, can be used as the final output of the drone flight path planning, guiding the drone to select a safe flight path, avoid high-risk areas, and ensure flight safety.

[0140] In order to facilitate the understanding of the present invention, some preferred embodiments of the present invention are further described below.

[0141] The method flow of the present invention provides a detailed solution from real-time data collection to final flight path decision in a typical application scenario of drone airspace monitoring and path planning. In this scenario, drone swarms are used to perform a series of tasks, and it is crucial to ensure their flight safety and avoid conflicts with other aircraft.

[0142] Step 1: The input module collects real-time motion parameter data from drones and aircraft, including position, velocity, and acceleration, as well as the timestamp of recording these data. These data are obtained through sensors on drones and aircraft and transmitted to the ground control station or data processing center. These sensors can include GPS modules, accelerometers, and gyroscopes to ensure the accuracy and real-time nature of the data.

[0143] Step 2: The sampling module converts the acquired data into a time series data set. Through the equally spaced sampling operation, a multidimensional data set containing motion parameters at continuous time points is formed. This data set provides the basis for subsequent data analysis and processing. The construction of the time series data set enables us to conduct a comprehensive analysis of the dynamic behavior of drones and aircraft, so as to better predict their motion trajectories and potential conflict risks.

[0144] Step 3: Perform data cleaning operations, including data segment selection, clustering, and Raida test to identify and remove outliers. Raida test operation calculates the residual error and standard deviation of the data and removes those data points with residual errors exceeding three times the standard deviation to obtain filtered data. Data cleaning is a critical step because it ensures that the data on which the analysis is based is high-quality and reliable, and avoids the influence of outliers on the analysis results.

[0145] Step 4: The screening module normalizes the screened data to eliminate the impact of different dimensions between features and improve the efficiency of data processing. The normalization operation scales the data to a uniform scale through a specific normalization function.

[0146] Step 5: The normalization module calculates the relative collision risk value between the drone and the aircraft based on the normalized data. By calculating the distance and acceleration vector modulus, the relative collision risk data is obtained, which provides a quantitative risk assessment for flight safety. By calculating the relative distance and relative speed between the two, possible collisions in the future can be predicted, so that preventive measures can be taken.

[0147] Step 6: The target point construction module uses the relative collision risk data and time series data set to construct a path planning matrix, extract the real-time location data of the drone, and plan the target point set data. Then, the risk information is embedded into the target point set data to form the risk target point set data. The path planning matrix can use a dynamic programming algorithm or an A* algorithm, combined with real-time updated collision risk data, to dynamically adjust the flight path of the drone to avoid potential collisions.

[0148] Step 7: The prediction module conducts a comprehensive information evaluation on the risk target point set data, calculates the comprehensive information evaluation value of each drone, and sorts the target point set based on this evaluation value to obtain the comprehensive information evaluation value. Through drone conflict detection, non-conflicting target points are screened out to ensure that the drone's flight path avoids high-risk areas, thereby ensuring flight safety. The comprehensive information evaluation value can be calculated based on multiple factors such as collision risk, path length, flight time, etc. to ensure that the drone chooses the optimal path.

[0149] The entire process improves the operational efficiency and safety of drone swarms through automated data processing and flight path planning, reduces the need for human intervention, and is able to adapt to changing airspace conditions. In this way, drone swarms can optimize flight paths, reduce operating costs, and increase the success rate of mission execution while complying with airspace rules. The application scenarios of this process include but are not limited to urban logistics distribution, agricultural monitoring, air traffic management, and emergency response missions, providing strong technical support for the widespread use of drones.

[0150] In summary, the present invention provides a data processing method based on drone airspace monitoring, characterized in that it includes: obtaining real-time motion parameter data and timestamps of each drone and aircraft in the drone group; performing data point equidistant sampling operations according to the motion parameter data and the timestamp to obtain a time series data set; performing data cleaning operations according to the time series data set to obtain screening data; performing normalization operations according to the screening data to obtain normalized data; performing relative collision risk calculations according to the normalized data to obtain relative collision risk data; performing path planning matrix construction operations according to the relative collision risk data and the time series data set to obtain risk target point set data; performing conflict-free screening operations according to the risk target point set data to obtain target points. The present invention reduces the risk of drone flight collisions by conducting real-time risk assessment during drone flight and embedding risk information into path points.

[0151] Reference Figure 2 An embodiment of the present invention provides a data processing system based on drone airspace monitoring, comprising:

[0152] An input module is used to obtain real-time motion parameter data and timestamps of each drone and aircraft in the drone swarm;

[0153] A sampling module, used for performing an equidistant sampling operation of data points according to the motion parameter data and the timestamp to obtain a time series data set;

[0154] A screening module, used for performing a data cleaning operation according to the time series data set to obtain screening data;

[0155] A normalization module, used to perform a normalization operation based on the screening data to obtain normalized data;

[0156] A risk assessment module, used to calculate the relative collision risk according to the normalized data to obtain relative collision risk data;

[0157] A target point construction module, used to perform a path planning matrix construction operation according to the relative collision risk data and the time series data set to obtain risk target point set data;

[0158] The output module is used to perform a conflict-free screening operation based on the risk target point set data to obtain the target point.

[0159] It should be noted that the data processing device based on drone airspace monitoring provided in an embodiment of the present invention is used to execute all the process steps of the data processing method based on drone airspace monitoring in the above-mentioned embodiment. The working principles and beneficial effects of the two correspond one to one, and therefore will not be repeated here.

[0160] The embodiment of the present invention further provides an electronic device. The electronic device includes: a processor, a memory, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, the steps in the above-mentioned data processing method embodiment based on drone airspace monitoring are implemented, such as Figure 1 Alternatively, the processor implements the functions of each module / unit in the above-mentioned device embodiments when executing the computer program.

[0161] Exemplarily, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to implement the present invention. The one or more modules / units may be a series of computer program instruction segments capable of implementing specific functions, which are used to describe the execution process of the computer program in the electronic device.

[0162] The electronic device may be a computing device such as a desktop computer, a notebook, a PDA, and a smart tablet. The electronic device may include, but is not limited to, a processor and a memory. Those skilled in the art will appreciate that the above components are merely examples of electronic devices and do not constitute a limitation on the electronic device. The electronic device may include more or fewer components than the above components, or may combine certain components, or different components. For example, the electronic device may also include input and output devices, network access devices, buses, etc.

[0163] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the electronic device, and uses various interfaces and lines to connect various parts of the entire electronic device.

[0164] The memory can be used to store the computer program and / or module, and the processor realizes various functions of the electronic device by running or executing the computer program and / or module stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0165] Wherein, if the module / unit integrated in the electronic device is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0166] It should be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the accompanying drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art may understand and implement it without paying any creative effort.

[0167] The specific embodiments described above further illustrate the purpose, technical solutions and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. It is particularly pointed out that for those skilled in the art, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of protection of the present invention.

Claims

1. A data processing method based on drone airspace monitoring, characterized in that: Executed by a computer, including: Obtain real-time motion parameter data and timestamps for each drone and aircraft in the drone swarm; According to the motion parameter data and the timestamp, a data point sampling operation with equal spacing is performed to obtain a time series data set; According to the time series data set, a data cleaning operation is performed to obtain screening data; Performing a normalization operation according to the screened data to obtain normalized data; performing relative collision risk calculation according to the normalized data to obtain relative collision risk data; Performing a path planning matrix construction operation according to the relative collision risk data and the time series data set to obtain risk target point set data; According to the risk target point set data, a conflict-free screening operation is performed to obtain the target point.

2. The data processing method based on drone airspace monitoring according to claim 1 is characterized in that: According to the motion parameter data and the timestamp, a data point equal-interval sampling operation is performed to obtain a time series data set, including: The motion parameter data includes position, velocity and acceleration; The time series dataset is constructed as follows: X={X0,X1,X2,...,X t ,…,X T } Among them, X t is a multidimensional dataset at time t; T represents the number of collected time series; Each X t It can be further expressed as: X t =[u1(t) u2(t) …u i (t)… u M (t) a1(t) a2(t) …a j (t)… a N (t)] T Among them, u i (t) is the motion parameter data of the i-th UAV at time t, a j (t) is the motion parameter data of the jth aircraft at time t, M is the total number of drones, and N is the total number of aircraft.

3. The data processing method based on drone airspace monitoring according to claim 1 is characterized in that: According to the time series data set, a data cleaning operation is performed to obtain screening data, including: According to the time series data set, a data segment selection operation is performed to obtain time segment data; Performing a clustering operation according to the time segment data to obtain time clustering data; According to the time clustering data, a Laida detection operation is performed to obtain screening data.

4. The data processing method based on drone airspace monitoring according to claim 3 is characterized in that: According to the time clustering data, a Laida detection operation is performed to obtain screening data, including: Performing a residual error calculation operation according to the time clustering data to obtain a residual error; Performing a standard deviation calculation operation according to the time clustering data to obtain a standard deviation; According to the residual error and the standard deviation, when the absolute value of the residual error is greater than three times the standard deviation, a deletion operation is performed to obtain screening data; The residual error calculation formula is as follows: Among them, v i represents the residual error of the i-th data; x i Represents the i-th data value; represents the arithmetic mean; The standard deviation calculation formula is as follows: Where σ represents the standard deviation; x i Represents the i-th data value; represents the arithmetic mean; n represents the total number of data.

5. The data processing method based on drone airspace monitoring according to claim 1 is characterized in that: According to the screening data, a normalization operation is performed to obtain normalized data, including: Perform feature selection based on the screening data to obtain similar feature data; Performing a feature normalization operation according to the similar feature data to obtain normalized data; The calculation formula of the feature normalization operation is as follows: X nnnn,k =norm(X k ) Among them, X k is the similar feature data of the drone group at time k, norm(X k ) is the normalization function, X norm,k Represents the normalized data of the drone swarm at time k.

6. The data processing method based on drone airspace monitoring according to claim 1 is characterized in that: According to the normalized data, relative collision risk calculation is performed to obtain relative collision risk data, including: Calculating the distance between the UAV and the aircraft based on the normalized data to obtain distance data; performing relative collision risk calculation according to the normalized data and the distance data to obtain relative collision risk data; The relative collision risk calculation formula is as follows: Among them, R ik represents the relative collision risk value between the i-th UAV and the k-th aircraft; d ik represents the distance value between the i-th UAV and the k-th aircraft; |u i | represents the magnitude of the acceleration vector of the i-th drone; |a k | represents the magnitude of the kth aircraft acceleration vector.

7. The data processing method based on drone airspace monitoring according to claim 1 is characterized in that: According to the relative collision risk data and the time series data set, a path planning matrix construction operation is performed to obtain risk target point set data, including: Extracting the drone location information based on the time series data set to obtain the drone real-time location data; According to the real-time position data of the UAV, path planning is performed to obtain target point set data; Performing a risk information embedding operation according to the target point set data and the relative collision risk data to obtain risk target point set data; The risk target point set data is as follows: P i ={p1,p2,…p j ,…,p m } Among them, P i represents the i-th drone risk target point set; p j represents the location data of the jth target point embedded with risk information; m represents the number of elements in the risk target point set.

8. The data processing method based on drone airspace monitoring according to claim 1 is characterized in that: According to the risk target point set data, a conflict-free screening operation is performed to obtain target points, including: Performing a comprehensive information evaluation based on the risk target point set data to obtain a comprehensive information evaluation value; Sorting the risk target point set data from small to large based on all the comprehensive information evaluation values ​​to obtain ordered target point set data; According to the ordered target point set data, sequentially selecting ordered target points as initial target points; A radius threshold is set. When it is determined that there are other drones within the radius of the initial target point, the initial target point is determined to be a drone collision point and the initial target point is discarded; when it is determined that there are no other drones within the radius of the initial target point, the initial target point is determined to be the target point; The calculation formula for comprehensive information evaluation is as follows: Among them, V i represents the comprehensive information evaluation value of the i-th UAV; N represents the total number of aircraft; R ik represents the relative collision risk value between the i-th UAV and the k-th aircraft; d ik Represents the distance value between the i-th UAV and the k-th aircraft.

9. A data processing system based on drone airspace monitoring, characterized in that: include: An input module is used to obtain real-time motion parameter data and timestamps of each drone and aircraft in the drone swarm; A sampling module, used for performing an equidistant sampling operation of data points according to the motion parameter data and the timestamp to obtain a time series data set; A screening module, used for performing a data cleaning operation according to the time series data set to obtain screening data; A normalization module, used to perform a normalization operation based on the screening data to obtain normalized data; A risk assessment module, used to calculate the relative collision risk according to the normalized data to obtain relative collision risk data; A target point construction module, used to perform a path planning matrix construction operation according to the relative collision risk data and the time series data set to obtain risk target point set data; The output module is used to perform a conflict-free screening operation based on the risk target point set data to obtain the target point.

10. An electronic device, characterized in that: It includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the data processing method based on drone airspace monitoring as described in any one of claims 1 to 8.

Images