Unmanned aerial vehicle automatic airport monitoring and maintenance system based on real-time data analysis

Abstract

The invention relates to the technical field of monitoring and maintenance, in particular to an unmanned aerial vehicle automatic airport monitoring and maintenance system based on real-time data analysis. The system comprises a detection analysis unit, an analysis prediction unit, a time prediction unit, a flight adjustment unit and a planning maintenance unit. According to the invention, the planning maintenance unit plans the wind-sheltering path from the position point of the unmanned aerial vehicle to the position point where the foreign matter rolls down in the airport runway according to the acquired wind power data, the acquired wind speed data and the acquired wind direction data, and then the unmanned aerial vehicle flies from the moment t according to the flight attitude of the unmanned aerial vehicle needing to be adjusted according to the planned wind-sheltering path. When the unmanned aerial vehicle flies to a position point where a foreign matter rolls down in an airport runway, automatic maintenance is carried out, when the unmanned aerial vehicle flies under the conditions of reduced wind power and optimized flight attitudes according to a planned wind sheltering path, a strong wind area can be effectively avoided, the flight stability is ensured, and the unmanned aerial vehicle is safe and reliable. And a foreign matter position point is reached in the shortest time, so that the searching and positioning time is shortened, and the airport runway maintenance efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of monitoring and maintenance, and in particular to an automatic airport monitoring and maintenance system using an unmanned aerial vehicle based on real-time data analysis. Background Art

[0002] Automatic airport monitoring and maintenance by drones refers to the use of drones and sensor technology to monitor and maintain airport infrastructure (such as runways). When there is wind at an airport, foreign objects may accidentally fall onto the runway under the action of the wind. Foreign objects may damage the aircraft wheels and landing gear, reducing the safety of the airport. The airport infrastructure is monitored and maintained by drone technology. The system formulates a flight path based on the location of the drone and the location of the foreign object. When the drone encounters a strong wind area during flight according to the formulated flight path, it is disturbed by the strong wind, causing the drone to deviate from the original flight path and the flight attitude of the drone (such as pitch and roll) is disrupted, making the drone more difficult to control, thereby reducing the stability of the drone's flight. At the same time, when the drone deviates from the original flight path, it needs to re-detect the location of the foreign object, which increases the search and positioning time and reduces the maintenance efficiency of the airport runway. Therefore, we provide an automatic airport monitoring and maintenance system for drones based on real-time data analysis. Summary of the invention

[0003] The purpose of the present invention is to provide an automatic airport monitoring and maintenance system for unmanned aerial vehicles based on real-time data analysis to solve the problems raised in the above-mentioned background technology.

[0004] To achieve the above-mentioned object, the present invention provides an automatic airport monitoring and maintenance system for unmanned aerial vehicles based on real-time data analysis, comprising a detection and analysis unit, an analysis and prediction unit, a time prediction unit, a flight adjustment unit, and a planning and maintenance unit; The detection and analysis unit monitors the airport runway infrastructure, wind force, wind speed and wind direction data in real time, obtains historical data and drone status data, detects foreign objects based on the airport runway infrastructure data monitored in real time, and then analyzes the rules based on the wind force, wind speed and wind direction data; The analysis and prediction unit is used to receive the data in the detection and analysis unit, analyze the rolling trajectory of the foreign object according to the historical data, and then predict the The location point where the foreign object rolled down the airport runway at the moment; The time prediction unit is used to receive data from the analysis prediction unit, and the time prediction unit takes data from the detection and analysis unit to predict The probability value of the wind speed decreasing on the airport runway at the moment; The flight adjustment unit is used to receive data from the detection and analysis unit, calculate the wind speed component in the horizontal axis direction and the wind speed component in the vertical axis direction, and then calculate the horizontal speed of the drone that needs to be adjusted and the vertical speed of the drone that needs to be adjusted according to the wind speed component in the horizontal axis direction and the wind speed component in the vertical axis direction, calculate the horizontal speed deviation of the drone and the vertical speed deviation of the drone using the horizontal speed and the vertical speed that need to be adjusted, and then derive the flight attitude of the drone that needs to be adjusted according to the horizontal speed deviation and the vertical speed deviation of the drone; The planning and maintenance unit is used to receive data from the time prediction unit, and the planning and maintenance unit takes data from the detection and analysis unit, the analysis and prediction unit, and the flight adjustment unit, and plans a wind avoidance path according to the taken data. The UAV moves from the predicted wind avoidance path according to the planned wind avoidance path. The drone's flight attitude can be adjusted as needed at all times, and the airport runway infrastructure can be automatically maintained.

[0005] As a further improvement of the technical solution, the detection and analysis unit includes a foreign body detection module and a regularity analysis module; The foreign object detection module uses the camera equipment equipped by the drone to monitor the airport runway infrastructure data in real time, and then obtains the wind force, wind speed and wind direction data of the airport runway in real time through the wind force, wind speed and wind direction sensors, and obtains historical data and drone status data. The background modeling technology is used to establish a background model based on the real-time monitored airport runway infrastructure data. , and then based on the real-time monitoring of airport runway infrastructure data and background models Detect foreign objects, locate the location of foreign objects through positioning technology, and record the rolling speed of foreign objects; The law analysis module is used to receive data from the foreign object detection module, and use a long short-term memory network to analyze the laws of wind force, wind speed and wind direction according to the acquired wind force, wind speed and wind direction data.

[0006] As a further improvement of the technical solution, the analysis and prediction unit includes a trajectory analysis module and a position prediction module; The trajectory analysis module is used to receive data from the law analysis module and the foreign object detection module, update the historical location points of the historical foreign objects and the foreign object location points into a historical location point data set, input them into the linear regression model for training to obtain corresponding multinomial regression coefficients, analyze the rolling trajectory of the foreign object according to the multinomial regression coefficients and the timestamps of the location points, and obtain the analyzed foreign object rolling trajectory ; in, refers to the nth multinomial regression coefficient, refers to the n-1th multinomial regression coefficient, refers to the first multinomial regression coefficient, refers to the intercept of the linear regression model, The n in refers to the timestamp index of the location point; The position prediction module is used to receive the foreign object rolling trajectory analyzed in the trajectory analysis module and the data in the foreign object detection module, and predict the foreign object according to the analyzed foreign object rolling trajectory, foreign object location point, foreign object rolling speed, location point timestamp and foreign object historical location point The point at which the foreign object rolled down the airport runway.

[0007] As a further improvement of the technical solution, the position prediction module predicts The location of the foreign object rolling down the airport runway at time step: Step 1: Collect historical locations of foreign objects , from the historical position of foreign objects Take the foreign body position point at time t-1 , Refers to the location of the foreign body The horizontal axis of time, Refers to the location of the foreign body The vertical coordinate of the time; Step 2: According to the analyzed foreign body rolling trajectory 、Foreign matter location point , the speed at which the foreign body rolls , the timestamp of the location point and the foreign body position at time t-1 predict The location of the foreign object rolling down the airport runway at the moment of the prediction The foreign body rolling position at the moment, the specific algorithm formula is: ; in, Refers to the horizontal coordinate of the foreign body position at time t, Refers to the ordinate of the foreign body position at time t, Refers to the horizontal velocity component of the foreign body, Refers to the vertical velocity component of the foreign body.

[0008] As a further improvement of the technical solution, the time prediction unit is used to receive the position prediction module At the moment when a foreign object rolls down on the airport runway, the time prediction unit takes data from the law analysis module, inputs the analyzed wind force law data, the analyzed wind speed law data and the analyzed wind direction law data into the logistic regression model for training, and the logistic regression model outputs the corresponding weight coefficient according to the training result, and predicts the wind force law data, the analyzed wind speed law data, the analyzed wind direction law data and the corresponding weight coefficient. The probability value of the wind speed decreasing on the airport runway at the moment is obtained, and the calculated probability value is obtained ;when Calculated probability value When If there is a decrease in wind speed on the airport runway at the moment, the decrease in wind speed is marked as 1; When the probability value is calculated When There is no wind reduction on the airport runway at the moment, and the absence of wind reduction is marked as 0.

[0009] As a further improvement of the technical solution, the flight adjustment unit includes a speed adjustment module and a posture adjustment module; The speed adjustment module is used to receive data from the foreign object detection module, obtain the drone speed from the drone status data, and then obtain the angle from the acquired wind direction data, and calculate the wind speed component in the horizontal axis direction according to the acquired wind speed data and angle. and the velocity component of the wind in the vertical direction Then, the horizontal speed that the drone needs to adjust is calculated based on the wind speed component in the horizontal direction, the wind speed component in the vertical direction and the drone speed. And the vertical speed that the drone needs to adjust ; The posture adjustment module is used to receive historical data from the foreign object detection module. And adjust the data in the speed module, calculate the horizontal speed deviation of the drone according to the horizontal speed that the drone needs to adjust, the vertical speed that the drone needs to adjust and the set expected speed of the drone Vertical speed deviation from the drone , according to the horizontal speed deviation of the UAV , UAV vertical speed deviation Based on historical data, the flight attitude of the drone that needs to be adjusted is obtained .

[0010] As a further improvement of the technical solution, the planning maintenance unit is used to receive the prediction in the time prediction unit At this moment, there is a wind reduction command on the airport runway. The planning and maintenance unit collects data from the foreign object detection module, the attitude adjustment module, and the position prediction module. The RRT algorithm is used to plan a wind avoidance path from the drone position to the position where the foreign object rolls on the airport runway according to the acquired wind force data, wind speed data, and wind direction data. The drone moves from the planned wind avoidance path to the location where the foreign object rolls on the airport runway. The drone's flight attitude is adjusted as needed at all times, and when the drone flies to the point where foreign objects roll down the airport runway, it automatically maintains the airport runway infrastructure.

[0011] Compared with the prior art, the present invention has the following beneficial effects: 1. In the UAV automatic airport monitoring and maintenance system based on real-time data analysis, the planning and maintenance unit plans a wind-avoiding path from the UAV position to the position where the foreign object rolls on the airport runway according to the acquired wind force data, wind speed data, and wind direction data. At this moment, the wind speed on the airport runway decreases, and the drone follows the planned wind-avoiding path. The drone's flight attitude is adjusted as needed at all times. When the drone flies to the point where foreign objects roll down on the airport runway, the airport runway infrastructure is automatically maintained. The drone flies according to the planned wind shelter path when the wind force is reduced and the drone's flight attitude is optimized. It can effectively avoid strong wind areas during the flight and prevent the drone from deviating from the path due to the influence of wind during flight, ensuring the stability of the flight. At the same time, it reaches the location of foreign objects in the shortest time to reduce search and positioning time and improve the maintenance efficiency of airport runways.

[0012] 2. In the UAV automatic airport monitoring and maintenance system based on real-time data analysis, the speed adjustment module calculates the wind speed component in the horizontal axis direction and the wind speed component in the vertical axis direction according to the acquired wind speed data and angle, and then calculates the horizontal speed and vertical speed that the UAV needs to adjust according to the wind speed component in the horizontal axis direction, the wind speed component in the vertical axis direction and the UAV speed. By calculating the wind speed component and adjusting the horizontal and vertical speeds of the UAV, the UAV can better offset the impact of wind on the flight path and maintain flight stability. BRIEF DESCRIPTION OF THE DRAWINGS

[0013] Figure 1 A unit block diagram of the present invention; Figure 2 It is a block diagram of a module unit of the present invention; Figure 3 It is a block diagram of the detection and analysis unit of the present invention; Figure 4 It is a block diagram of the analysis and prediction unit of the present invention; Figure 5 This is a block diagram of the flight adjustment unit of the present invention.

[0014] The meaning of each number in the figure is: 1. Detection and analysis unit; 11. Foreign body detection module; 12. Regularity analysis module; 2. Analysis and prediction unit; 21. Trajectory analysis module; 22. Position prediction module; 3. Time prediction unit; 4. Flight adjustment unit; 41. Speed ​​adjustment module; 42. Attitude adjustment module; 5. Planning and maintenance unit. DETAILED DESCRIPTION

[0015] The following will be combined with the accompanying drawings in 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. Example

[0016] The present invention provides an automatic airport monitoring and maintenance system based on real-time data analysis by unmanned aerial vehicles. Figure 1 , including a detection and analysis unit 1, an analysis and prediction unit 2, a time prediction unit 3, a flight adjustment unit 4 and a planning and maintenance unit 5; The detection and analysis unit 1 monitors the airport runway infrastructure, wind force, wind speed and wind direction data in real time, and obtains historical data and drone status data, detects foreign objects based on the real-time monitored airport runway infrastructure data, and then analyzes the rules based on the wind force, wind speed and wind direction data; the analysis and prediction unit 2 is used to receive the data in the detection and analysis unit 1, analyze the rolling trajectory of the foreign object based on the historical data, and then predict the foreign object based on the analyzed rolling trajectory of the foreign object and the historical data. The location point where the foreign object rolls down on the airport runway at the moment; the time prediction unit 3 is used to receive the data in the analysis and prediction unit 2, and the time prediction unit 3 takes data from the detection and analysis unit 1 to predict The probability value of the wind force decreasing on the airport runway at the moment; the flight adjustment unit 4 is used to receive the data in the detection and analysis unit 1, calculate the speed component of the wind in the horizontal axis direction and the speed component of the wind in the vertical axis direction, and then calculate the horizontal speed of the drone that needs to be adjusted and the vertical speed of the drone that needs to be adjusted according to the speed component of the wind in the horizontal axis direction and the speed component of the wind in the vertical axis direction, calculate the horizontal speed deviation of the drone and the vertical speed deviation of the drone using the horizontal speed that needs to be adjusted and the vertical speed that needs to be adjusted, and then derive the flight attitude of the drone that needs to be adjusted according to the horizontal speed deviation of the drone and the vertical speed deviation of the drone; the planning and maintenance unit 5 is used to receive the data in the time prediction unit 3, the planning and maintenance unit 5 takes data from the detection and analysis unit 1, the analysis and prediction unit 2 and the flight adjustment unit 4, and plans a wind shelter path according to the taken data, and the drone moves from the predicted wind shelter path according to the planned wind shelter path. The drone's flight attitude can be adjusted as needed at all times, and the airport runway infrastructure can be automatically maintained.

[0017] The following is a refinement of the above units, see Figure 2-Figure 5 The detection and analysis unit 1 includes a foreign body detection module 11 and a regularity analysis module 12; Since foreign objects may fall from the airport runway under the action of wind, which may cause direct damage to the aircraft wheels and landing gear, resulting in reduced airport safety, the foreign object detection module 11 uses the camera equipment equipped by the drone to monitor the airport runway infrastructure data in real time. , and then use wind force, wind speed and wind direction sensors to obtain real-time wind force, wind speed and wind direction data of the airport runway, and obtain historical data and drone status data , using background modeling technology based on real-time monitoring of airport runway infrastructure data Building a background model , and then based on the real-time monitoring of airport runway infrastructure data and background model Detect foreign objects and locate the location of foreign objects through positioning technology , while recording the speed of the foreign body rolling ; Historical data Including historical locations of foreign objects , the timestamp of the location point (refers to the time difference between the previous location of the foreign object and the current location of the foreign object), pitch angle proportional coefficient and roll angle proportionality factor ; Drone status data Including drone location points , UAV speed ; Steps to detect foreign matter: Step 1: Collect real-time monitoring data of airport runway infrastructure and divided into pixel areas and record the grayscale value of each pixel area ; Step ②: Based on real-time monitoring of airport runway infrastructure data , pixel area and grayscale value of the area pixel Calculate pixel position The average gray value , the specific algorithm formula is: ; in, refers to The location area Grayscale value of the pixel; Step 3: Location The average gray value , Background model and set thresholds Detect the presence of foreign objects in the airport runway infrastructure monitored in real time. When the location of the airport runway infrastructure is detected in real time There is foreign matter; The law analysis module 12 is used to receive the wind data obtained by the foreign object detection module 11 , wind speed data obtained And the wind direction data obtained , using the long short-term memory network based on the acquired wind data , wind speed data obtained And the wind direction data obtained Analyze the laws of wind force, wind speed and wind direction separately. Changes in wind speed and direction can affect the rolling of foreign objects on the runway. Early analysis and monitoring can allow for timely cleaning and maintenance, thereby improving the maintenance efficiency of airport runways.

[0018] The analysis and prediction unit 2 includes a trajectory analysis module 21 and a position prediction module 22; The trajectory analysis module 21 is used to receive the wind force law data analyzed in the law analysis module 12 , analyzed wind speed pattern data and analyzed wind direction data , receiving the historical location points of the historical foreign objects in the foreign object detection module 11 、Foreign matter location point and the timestamp of the location , the historical location of historical foreign objects Update the foreign object location points into the historical location point data set , the historical location point dataset Input the linear regression model for training and output the corresponding multinomial regression coefficients , according to the multinomial regression coefficients and the timestamp of the location Analyze the rolling trajectory of foreign bodies and obtain the analyzed rolling trajectory of foreign bodies By accurately analyzing the rolling trajectory of foreign objects, the number of foreign objects on the airport runway can be reduced, reducing the possibility of aircraft encountering risks during takeoff and landing; in, refers to the nth multinomial regression coefficient, refers to the n-1th multinomial regression coefficient, refers to the first multinomial regression coefficient, refers to the intercept (constant term) of the linear regression model, The n in refers to the timestamp index of the location point; The position prediction module 22 is used to receive the foreign body rolling trajectory analyzed by the trajectory analysis module 21 and the foreign body position point in the foreign body detection module 11 , the speed at which the foreign body rolls , the timestamp of the location point 、The historical location of foreign objects , according to the analysis of the foreign body rolling trajectory 、Foreign matter location point , the speed at which the foreign body rolls , the timestamp of the location point and historical locations of foreign objects predict By predicting the location of foreign objects rolling down the airport runway at any moment, the drone can respond quickly and issue an early warning, reducing the time from discovering foreign objects to taking action, improving response speed, and The position point command of the foreign object rolling down on the airport runway at the moment is transmitted to the time prediction unit 3; The position prediction module 22 predicts The location of the foreign object rolling down the airport runway at time step: Step 1: Collect historical locations of foreign objects , from the historical position of foreign objects Take the foreign body position point at time t-1 , Refers to the location of the foreign body The horizontal axis of time, Refers to the location of the foreign body The vertical coordinate of the time; Step 2: According to the analyzed foreign body rolling trajectory 、Foreign matter location point , the speed at which the foreign body rolls , the timestamp of the location point and the foreign body position at time t-1 predict The location of the foreign object rolling down the airport runway at the moment of the prediction The foreign body rolling position at the moment, the specific algorithm formula is: ; in, Refers to the horizontal coordinate of the foreign body position at time t, Refers to the ordinate of the foreign body position at time t, Refers to the horizontal velocity component of the foreign body, Refers to the vertical velocity component of foreign objects. Accurate prediction enables drones to automatically intercept and maintain foreign objects before they reach critical areas (such as airport runways), avoiding potential safety risks on airport runways and improving the safety of airport runways. The time prediction unit 3 is used to receive the position prediction module 22 At the moment when the foreign object is rolling down the airport runway, the time prediction unit 3 sends the wind force law data analyzed by the law analysis module 12 to the time prediction unit 3. , analyzed wind speed pattern data and analyzed wind direction data , the wind force law data analyzed , analyzed wind speed pattern data and analyzed wind direction data Input the logistic regression model for training, and the logistic regression model outputs the corresponding weight coefficient according to the training results , according to the wind force law data analyzed , analyzed wind speed pattern data , analyzed wind direction data and the corresponding weight coefficients predict The probability value of the wind speed decreasing on the airport runway at the moment is obtained, and the calculated probability value is obtained , using the predicted data of reduced wind speed, the drone can quickly complete the detection and cleaning of runway foreign objects within the optimal time window, thereby shortening the overall operation time and improving the operation efficiency of the drone. When At this moment, there is a decrease in wind speed on the airport runway, and then the forecast At the moment, there is a wind reduction command on the airport runway, which is transmitted to the planning and maintenance unit 5, and the wind reduction is marked as 1. When the calculated probability value When If there is no wind reduction on the airport runway at the moment, the absence of wind reduction is marked as 0; predict The realization principle of the probability value of the wind force decreasing on the airport runway at the moment: Collect and analyze wind pattern data , analyzed wind speed pattern data , analyzed wind direction data and the corresponding weight coefficients predict The probability value of the wind speed decreasing on the airport runway at the moment is obtained, and the calculated probability value is obtained , the specific algorithm formula is: ; in, refers to the intercept (constant term) of the logistic regression model, Refers to the first weight coefficient , Refers to the second weight coefficient , Refers to the third weight coefficient , which is used to predict The probability value of the wind speed decreasing on the airport runway at the moment. By predicting the probability of wind speed decreasing, the drone can perform runway foreign body maintenance operations during the period of low wind speed, thereby reducing the impact of strong wind on the stability of the drone. The flight adjustment unit 4 includes a speed adjustment module 41 and an attitude adjustment module 42; The speed adjustment module 41 is used to receive the drone status data from the foreign object detection module 11. , wind data obtained , wind speed data obtained And the wind direction data obtained , from the drone status data Take the drone speed , and then from the wind direction data obtained Take an angle , based on the wind speed data obtained and angle Calculate the wind velocity component in the horizontal direction and the velocity component of the wind in the vertical direction , and then according to the wind speed component in the horizontal direction , the velocity component of the wind in the longitudinal direction and drone speed Calculate the horizontal speed that the drone needs to adjust And the vertical speed that the drone needs to adjust , by calculating the wind’s velocity components and adjusting the drone’s horizontal and vertical speeds, the drone can better offset the wind’s impact on the flight path and maintain flight stability; The posture adjustment module 42 is used to receive the historical data in the foreign body detection module 11 And the horizontal speed that the drone needs to adjust in the speed adjustment module 41 , the vertical speed that the drone needs to adjust , adjust the horizontal speed according to the needs of the drone , the vertical speed that the drone needs to adjust and the desired speed of the drone Calculate the horizontal velocity deviation of the drone Vertical speed deviation from the drone , and then from historical data The pitch angle ratio coefficient is taken in and roll angle proportionality factor , according to the horizontal speed deviation of the UAV , UAV vertical speed deviation , pitch angle proportional coefficient and roll angle proportionality factor Get the UAV flight attitude that needs to be adjusted ,Timely adjustment of pitch and roll angles can reduce the instability caused by horizontal and vertical speed deviations, ensuring that the UAV maintains stable flight under changing meteorological conditions; Calculate the UAV flight attitude that needs to be adjusted Implementation steps: Step 1: Use the pitch angle ratio coefficient Vertical speed deviation from the drone Calculate the pitch angle that the drone needs to adjust ; Step 2: Use the roll angle proportionality factor The horizontal speed deviation from the drone Calculate the roll angle that the drone needs to adjust ; Step 3: Adjust the pitch angle according to the needs of the drone And the roll angle that the drone needs to adjust Get the UAV flight attitude that needs to be adjusted ; The planning maintenance unit 5 is used to receive the prediction in the time prediction unit 3 At the moment there is a wind reduction command on the airport runway, the planning and maintenance unit 5 takes the acquired wind data to the foreign object detection module 11 , wind speed data obtained , wind direction data obtained 、UAV location point and the flight attitude of the drone that needs to be adjusted in the attitude adjustment module 42 The planning and maintenance unit 5 takes the location point where the foreign object rolls down the airport runway from the location prediction module 22, and uses the RRT algorithm to calculate the location point based on the acquired wind data. , wind speed data obtained , wind direction data obtained Planning from UAV location point The wind-avoiding path to the point where the foreign object rolled down the airport runway is predicted At this moment, the wind speed on the airport runway decreases, and the drone follows the planned wind-avoiding path. Adjust the drone's flight posture as needed at any time When the UAV flies to the location where foreign objects roll down on the airport runway, it automatically maintains the airport runway infrastructure. When the UAV flies according to the planned wind avoidance path when the wind force is reduced and the UAV flight posture is optimized, it can effectively avoid strong wind areas during the flight, especially when the wind force is reduced, to prevent the UAV from deviating from the path under the influence of the wind during the flight, ensure the stability of the flight, and reach the location of the foreign object in the shortest time, reducing the search and positioning time and improving the maintenance efficiency of the airport runway.

[0019] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments. The above embodiments and descriptions are only preferred examples of the present invention and are not intended to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, which fall within the scope of the present invention. The scope of protection of the present invention is defined by the attached claims and their equivalents.

Claims

1. UAV automatic airport monitoring and maintenance system based on real-time data analysis, characterized by: It includes a detection and analysis unit (1), an analysis and prediction unit (2), a time prediction unit (3), a flight adjustment unit (4), and a planning and maintenance unit (5); The detection and analysis unit (1) monitors the airport runway infrastructure, wind force, wind speed and wind direction data in real time, obtains historical data and drone status data, detects foreign objects based on the airport runway infrastructure data monitored in real time, and then analyzes the rules based on the wind force, wind speed and wind direction data; The analysis and prediction unit (2) is used to receive the historical data, wind force, wind speed and wind direction data in the detection and analysis unit (1), analyze the rolling trajectory of the foreign object according to the historical data, and then analyze the laws of wind force, wind speed and wind direction according to the wind force, wind speed and wind direction data, and use the analyzed foreign object rolling trajectory and historical data to predict the At the moment when the foreign object rolls down on the airport runway, a command of the location where the foreign object rolls down is transmitted to the time prediction unit (3); The time prediction unit (3) is used to receive a command of a position point where a rollover occurs in the analysis and prediction unit (2), and the time prediction unit (3) takes the wind force, wind speed and wind direction regularity data analyzed in the detection and analysis unit (1) to predict the The probability value of the wind speed decreasing on the airport runway at the moment, and the wind speed decreasing command is transmitted to the planning and maintenance unit (5); The flight adjustment unit (4) is used to receive the drone status data, wind speed and wind direction data in the detection and analysis unit (1), calculate the wind speed component in the horizontal axis direction and the wind speed component in the vertical axis direction according to the wind speed and wind direction data, and then calculate the horizontal speed of the drone that needs to be adjusted and the vertical speed of the drone that needs to be adjusted according to the wind speed component in the horizontal axis direction, the wind speed component in the vertical axis direction and the drone status data, calculate the horizontal speed deviation of the drone and the vertical speed deviation of the drone using the horizontal speed and the vertical speed, and then derive the flight attitude of the drone that needs to be adjusted according to the horizontal speed deviation and the vertical speed deviation of the drone; The planning and maintenance unit (5) is used to receive a wind force reduction command from the time prediction unit (3), the planning and maintenance unit (5) obtains data from the detection and analysis unit (1), the analysis and prediction unit (2) and the flight adjustment unit (4), plans a wind avoidance path based on the obtained data, and the UAV moves from the predicted wind avoidance path according to the planned wind avoidance path. The drone's flight attitude can be adjusted as needed at all times, and the airport runway infrastructure can be automatically maintained.

2. The UAV automatic airport monitoring and maintenance system based on real-time data analysis according to claim 1 is characterized by: The detection and analysis unit (1) comprises a foreign matter detection module (11) and a regularity analysis module (12); The foreign object detection module (11) uses the camera equipment equipped by the drone to monitor the airport runway infrastructure data in real time, and then obtains the wind force, wind speed and wind direction data of the airport runway in real time through the wind force, wind speed and wind direction sensors, and obtains historical data and drone status data, and uses background modeling technology to establish a background model based on the real-time monitored airport runway infrastructure data. , and then based on the real-time monitoring of airport runway infrastructure data and background models Detect foreign objects, locate the location of foreign objects through positioning technology, and record the rolling speed of foreign objects; The law analysis module (12) is used to receive data from the foreign object detection module (11), and to use a long short-term memory network to analyze the laws of wind force, wind speed and wind direction based on the acquired wind force, wind speed and wind direction data.

3. The UAV automatic airport monitoring and maintenance system based on real-time data analysis according to claim 2 is characterized by: The analysis and prediction unit (2) comprises a trajectory analysis module (21) and a position prediction module (22); The trajectory analysis module (21) is used to receive data from the regularity analysis module (12) and the foreign object detection module (11), update the historical location points of the historical foreign objects and the foreign object location points into a historical location point data set, input them into a linear regression model for training to obtain corresponding multinomial regression coefficients, analyze the rolling trajectory of the foreign object based on the multinomial regression coefficients and the timestamps of the location points, and obtain the analyzed foreign object rolling trajectory ; in, refers to the nth multinomial regression coefficient, refers to the n-1th multinomial regression coefficient, refers to the first multinomial regression coefficient, refers to the intercept of the linear regression model, The n in refers to the timestamp index of the location point; The position prediction module (22) is used to receive the foreign object rolling trajectory analyzed in the trajectory analysis module (21) and the data in the foreign object detection module (11), and predict the foreign object according to the analyzed foreign object rolling trajectory, the foreign object location point, the foreign object rolling speed, the time stamp of the location point and the historical location point of the foreign object. The point at which the foreign object rolled down the airport runway.

4. The UAV automatic airport monitoring and maintenance system based on real-time data analysis according to claim 3 is characterized by: The position prediction module (22) predicts The location of the foreign object rolling down the airport runway at time step: Step 1: Collect historical locations of foreign objects , from the historical position of foreign objects Take the foreign body position point at time t-1 , Refers to the location of the foreign body The horizontal axis of time, Refers to the location of the foreign body The vertical coordinate of the time; Step 2: According to the analyzed foreign body rolling trajectory 、Foreign matter location point , the speed at which the foreign body rolls , the timestamp of the location point and the foreign body position at time t-1 predict The location of the foreign object rolling down the airport runway at the moment of the prediction The foreign body rolling position at the moment, the specific algorithm formula is: ； in, Refers to the horizontal coordinate of the foreign body position at time t, Refers to the ordinate of the foreign body position at time t, Refers to the horizontal velocity component of the foreign body, Refers to the vertical velocity component of the foreign body.

5. The UAV automatic airport monitoring and maintenance system based on real-time data analysis according to claim 3 is characterized by: The time prediction unit (3) is used to receive the position prediction module (22) At the moment when the foreign object rolls down on the airport runway, the time prediction unit (3) takes data from the law analysis module (12), inputs the analyzed wind force law data, the analyzed wind speed law data and the analyzed wind direction law data into the logistic regression model for training, and the logistic regression model outputs the corresponding weight coefficient according to the training result, and predicts the wind force law data, the analyzed wind speed law data, the analyzed wind direction law data and the corresponding weight coefficient according to the analyzed wind force law data, the analyzed wind speed law data, the analyzed wind direction law data and the corresponding weight coefficient. The probability value of the wind speed decreasing on the airport runway at the moment is obtained, and the calculated probability value is obtained ; When the probability value is calculated When If there is a decrease in wind speed on the airport runway at the moment, the decrease in wind speed is marked as 1; When the probability value is calculated When There is no wind reduction on the airport runway at the moment, and the absence of wind reduction is marked as 0.

6. The UAV automatic airport monitoring and maintenance system based on real-time data analysis according to claim 2 is characterized by: The flight adjustment unit (4) comprises a speed adjustment module (41) and a posture adjustment module (42); The speed adjustment module (41) is used to receive data from the foreign object detection module (11), obtain the speed of the drone from the drone status data, and then obtain the angle from the acquired wind direction data, and calculate the speed component of the wind in the horizontal axis direction according to the acquired wind speed data and angle. and the velocity component of the wind in the vertical direction Then, the horizontal speed that the drone needs to adjust is calculated based on the wind speed component in the horizontal direction, the wind speed component in the vertical direction and the drone speed. And the vertical speed that the drone needs to adjust ; The posture adjustment module (42) is used to receive historical data from the foreign object detection module (11). and the data in the speed adjustment module (41), and calculating the horizontal speed deviation of the drone according to the horizontal speed that the drone needs to adjust, the vertical speed that the drone needs to adjust, and the set expected speed of the drone. Vertical speed deviation from the drone , according to the horizontal speed deviation of the UAV , UAV vertical speed deviation Based on historical data, the flight attitude of the drone that needs to be adjusted is obtained .

7. The UAV automatic airport monitoring and maintenance system based on real-time data analysis according to claim 5 is characterized by: The planning maintenance unit (5) is used to receive the time prediction unit (3) At the moment, there is a wind reduction command on the airport runway. The planning and maintenance unit (5) obtains data from the foreign object detection module (11), the attitude adjustment module (42) and the position prediction module (22). The RRT algorithm is used to plan a wind avoidance path from the position of the UAV to the position where the foreign object rolls on the airport runway according to the acquired wind force data, the acquired wind speed data and the acquired wind direction data. The UAV moves from the position of the foreign object to the airport runway according to the planned wind avoidance path. The drone's flight attitude is adjusted as needed at all times, and when the drone flies to the point where foreign objects roll down the airport runway, it automatically maintains the airport runway infrastructure.

Images