Unmanned aerial vehicle flight integrated management method and system

Abstract

The invention provides an unmanned aerial vehicle flight integrated management method and system, and relates to the technical field of unmanned aerial vehicles, and the method comprises the steps: obtaining task information of a target unmanned aerial vehicle, and determining a first route according to the task information; optimizing the first route to obtain an optimized route; and monitoring the flight condition of the target unmanned aerial vehicle flying according to the optimized route in real time, and responding to a corresponding strategy to perform exception handling when an exception is detected. The method comprises the following steps: optimizing a first route determined according to task information of a target unmanned aerial vehicle to obtain an optimized route; and the flight condition of the target unmanned aerial vehicle flying according to the optimized route is monitored in real time, and when abnormity is detected, the corresponding strategy is responded to carry out abnormity processing, so that the flight safety and the flight efficiency of the unmanned aerial vehicle can be effectively enhanced, and the task execution efficiency is further improved.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicles, and in particular to a method and system for integrated management of unmanned aerial vehicle flight. Background Art

[0002] In recent years, with the rapid development of drone technology, drones have been widely used in many fields such as aerial photography, environmental monitoring, and target tracking. In actual flight, drones face complex flight environments, which puts forward higher requirements for flight safety and efficiency. However, most of the drone flight management methods in the existing technology have problems such as slow response and poor adaptability, which are difficult to meet the growing flight needs.

[0003] Therefore, the present invention provides a method and system for integrated management of UAV flight, which are used to effectively enhance the flight safety and efficiency of UAV. Summary of the invention

[0004] The present invention provides a method and system for integrated management of unmanned aerial vehicle (UAV) flight, which is used to optimize a first route determined according to mission information of a target UAV to obtain an optimized route; monitor the flight status of the target UAV flying along the optimized route in real time, and when an abnormality is detected, respond to a corresponding strategy to handle the abnormality, which can effectively enhance the flight safety and efficiency of the UAV, thereby improving the execution efficiency of the mission.

[0005] The present invention provides a method for integrated management of unmanned aerial vehicle flight, comprising: Step 1: Obtain and determine the first route based on the mission information of the target UAV; Step 2: Optimize the first route to obtain an optimized route; Step 3: Monitor the flight status of the target UAV according to the optimized route in real time, and when an abnormality is detected, respond to the corresponding strategy to handle the abnormality.

[0006] Preferably, obtaining and determining the first route according to the mission information of the target UAV includes: Extracting a first starting point and a first end point from the mission information of the target UAV; Using the first starting point and the first end point as screening conditions, screening out historical flight routes that are consistent with the first starting point and the first end point from the drone record database; Determine the mission type based on the mission information of the target UAV; According to the mission type, the acquired historical flight routes are analyzed and screened to obtain the first route.

[0007] Preferably, according to the mission type, the acquired historical flight routes are analyzed and screened to obtain a first route, including: When the mission type is a single mission type, if there is a single historical flight route, the current historical flight route is output as the first route; If there are multiple historical flight routes, the adaptation coefficient of each currently acquired historical flight route is calculated; The calculation formula of the adaptation coefficient is as follows: ; In the formula, Expressed as the adaptation coefficient of the i-th historical flight route; It is represented as the route length of the i-th historical flight route; It is expressed as the maximum flight speed of the current target UAV; It is represented by the influence weight of the estimated flight time of the UAV's flight route on the evaluation of route adaptation; It represents the maximum flight altitude of the current target UAV; It is represented as the maximum altitude of the i-th historical flight route; Expressed as the standard deviation of the altitude change of the i-th historical flight route; It is expressed as the weight of the impact of the altitude change of the flight route on the evaluation of route adaptation; It is expressed as the evaluation coefficient of avoiding restricted areas of the ith historical flight route; It is the weight of the ability to avoid restricted areas on the evaluation of route adaptation; Select the historical flight route with the highest adaptation coefficient as the first route output; If there is no historical flight route, a default route that matches the key matching parameters of the current target UAV is used as the first route output; When the mission type is a collaborative mission type, if there is a single historical flight route and the priority of the current target UAV is the first, the currently acquired historical flight route is regarded as the first route; If the priority of the current target UAV is not the first, a default route that matches the key matching parameters of the current target UAV is used as the first route output; If there is no historical flight route, a default route that matches the key matching parameters of the current target UAV is used as the first route output; If there are multiple historical flight routes, the historical flight route with an adaptation coefficient greater than the set adaptation threshold is regarded as the first candidate route; If there is a first candidate route, determining a first starting time for the current target UAV to travel the current first candidate route based on the priority order of the target UAVs; Based on the first starting time and the minimum driving speed of the target UAV, determine the first predicted time when the current target UAV completes the mission by traveling the current first candidate route; If the first predicted moments corresponding to all the first candidate routes obtained exceed the mission period of the current target UAV, a default route that matches the key matching parameters of the current target UAV is used as the first route output; If there is a first predicted time corresponding to the first candidate route that does not exceed the mission period of the current target UAV, the corresponding first candidate route is regarded as a candidate route; When there is only a single candidate route, the current candidate route is output as the first route; When there are multiple candidate routes, the candidate route with the highest fitness coefficient with the current target UAV is selected as the first route output.

[0008] Preferably, the mission information includes takeoff start and end points, mission type, mission deadline, mission content, priority order, and a priority table for assigning UAV tasks.

[0009] Preferably, optimizing the first route to obtain an optimized route includes: Traversing each sub-route in the first route to determine a first angle between adjacent sub-route; When there is a first angle greater than a set angle threshold, the corresponding sub-route is marked as a sharp turn segment; If there is only a single sharp turn segment, the current sharp turn segment is smoothed to obtain a new sub-route; Evaluating the new sub-route based on flight reliability to obtain a reliability evaluation coefficient; If the reliability evaluation coefficient is greater than the set evaluation threshold, the current route is determined to be reliable, and the corresponding sharp turn section is replaced with a new sub-route to generate an optimized route; If the reliability evaluation coefficient is not greater than the set evaluation threshold, the current route is determined to be unreliable, and the current sharp turn segment is smoothed again until a reliable new sub-route is obtained to replace the sharp turn segment and generate an optimized route; When there are multiple sharp turn segments, and the sharp turn segments are continuous, the first and last points of the continuous sharp turn segments are extracted and connected in a straight line to generate a first preset route; Evaluating the first preset route based on flight reliability to obtain a reliability evaluation coefficient; If the reliability evaluation coefficient is greater than the set evaluation threshold, the current route is determined to be reliable, and the first preset route is used to replace the corresponding continuous sharp turn segments to generate an optimized route; If the reliability evaluation coefficient is not greater than the set evaluation threshold, the current route is determined to be unreliable. At this time, the sharp turns with continuous relationships are smoothed in turn, and the reliable route replaces the corresponding sharp turns with continuous relationships, and finally generates an optimized route; A sharp turn segment that is not continuous with other sharp turn segments is marked as a first sharp turn segment and smoothed to obtain a second preset route; The effect of the second preset route is verified based on flight reliability. If the reliability evaluation coefficient is greater than a set evaluation threshold, the current route is determined to be reliable, and the corresponding first sharp turn section is replaced with the second preset route to generate an optimized route. If the reliability evaluation coefficient is not greater than the set evaluation threshold, the current route is determined to be unreliable, and the corresponding first sharp turn segment is smoothed again until a reliable second preset route is obtained to replace the first sharp turn segment, thereby generating an optimized route.

[0010] Preferably, the calculation formula of the reliability evaluation coefficient is as follows: ; In the formula, Y0 represents the reliability evaluation coefficient of the current route; It is expressed as the performance constraint satisfaction coefficient of the UAV on the current route; It is expressed as the weight of the impact of the performance constraint satisfaction capability on the reliability of the current route; It is expressed as the route length of the original sharp turn section; It is expressed as the route length of the current route; It is expressed as the weight of the impact of the change in route length on the reliability of the current route; It represents the set safety distance threshold between the flight route and the no-fly zone; It is represented by the distance between the jth no-fly zone and the current route that does not maintain a safe distance from the current route, where j=1, 2, 3, , n1; It is expressed as the total number of no-fly zones that do not maintain a safe distance from the current route; It indicates the set safety distance threshold between the route and the restricted area; It is represented by the distance between the kth no-fly zone and the current route that does not maintain a safe distance from the current route, where k=1, 2, 3, , n2; It is expressed as the total number of restricted areas that do not maintain a safe distance from the current route; It is expressed as the weight of the ability to avoid restricted areas on the reliability of the current route.

[0011] Preferably, real-time monitoring of the flight status of the target UAV according to the optimized route and, when an abnormality is detected, responding to a corresponding strategy to handle the abnormality include: Obtaining the initial battery power of the target UAV; When the initial battery power is greater than the set battery power, the target UAV flies according to the optimized route; When the initial battery power is not greater than the set battery power, the target drone sends a low-battery signal to the maintenance personnel; The maintenance personnel receive the low-battery signal and replace the battery of the current target drone; Using a set data collection tool to collect in real time the flight status of the target UAV flying along the optimized route to obtain first real-time flight data; Inputting the first real-time flight data into an anomaly detection model established in advance based on a machine learning algorithm to obtain a flight anomaly detection result; According to the flight anomaly detection result, if there is a flight anomaly type, sending an alarm signal to a ground control station; For abnormal situations where the flight abnormality type is the presence of dynamic obstacles, real-time dynamic obstacle avoidance operations are performed based on the real-time flight environment data of the target UAV; For anomalies where the flight anomaly type is not the presence of a dynamic obstacle, an anomaly handling strategy matching the current flight anomaly type is extracted from the set anomaly type-strategy mapping table to perform an anomaly handling operation; The exception handling process is recorded, a flight exception maintenance report is generated and stored in the flight exception record library.

[0012] Preferably, for an abnormal situation in which the flight abnormality type is the presence of a dynamic obstacle, a real-time dynamic obstacle avoidance operation based on the real-time flight environment data of the target UAV includes: After filtering, denoising and motion compensation of the real-time flight environment data of the target UAV, a first event image is obtained; Performing dynamic target segmentation on the first event image to obtain a first detection result; Performing dynamic obstacle detection on the image collected by the RGB camera to obtain a second detection result; Fusing the first detection result and the second detection result to obtain a target detection result; Combining the target detection result with a depth camera, performing depth information analysis on the dynamic obstacle, and obtaining first depth information; The first depth information is used as an input quantity, input into a Kalman filter and a time delay is introduced, and a first position and a first speed of the current dynamic obstacle are output; Determining a first relative speed according to the first speed and the real-time speed of the current target UAV; Based on the first relative speed, determining a first speed obstacle area of ​​the current target UAV; Inputting the first position, the first speed, and the first speed obstacle area of ​​the current dynamic obstacle into the strategy generation model constructed by the pre-trained neural network, and outputting a first obstacle avoidance strategy; The target UAV executes the first obstacle avoidance strategy and performs dynamic obstacle avoidance operations.

[0013] The present invention provides an unmanned aerial vehicle flight integrated management system, comprising: Flight route acquisition module: used to obtain and determine the first route based on the mission information of the target UAV; A flight route optimization module: used to optimize the first route to obtain an optimized route; UAV flight monitoring module: used to monitor the flight status of the target UAV according to the optimized route in real time, and when an abnormality is detected, respond to the corresponding strategy for abnormal processing.

[0014] Compared with the prior art, the present invention has the following beneficial effects: The optimized route is obtained by optimizing the first route determined according to the mission information of the target UAV; the flight status of the target UAV flying along the optimized route is monitored in real time, and when an abnormality is detected, the corresponding strategy is responded to for abnormal processing, which can effectively enhance the flight safety and efficiency of the UAV, thereby improving the execution efficiency of the mission.

[0015] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings.

[0016] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings: Figure 1 This is a flow chart of a method for integrated management of UAV flight in an embodiment of the present invention; Figure 2 This is a structural diagram of an integrated management system for unmanned aerial vehicle flight in an embodiment of the present invention. DETAILED DESCRIPTION

[0018] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0019] The embodiment of the present invention provides a method for integrated management of unmanned aerial vehicle flight. Figure 1 As shown, including: Step 1: Obtain and determine the first route based on the mission information of the target UAV; Step 2: Optimize the first route to obtain an optimized route; Step 3: Monitor the flight status of the target UAV according to the optimized route in real time, and when an abnormality is detected, respond to the corresponding strategy to handle the abnormality.

[0020] In this embodiment, a drone refers to an unmanned aircraft, which is an unmanned aerial vehicle controlled by a radio remote control device or a self-programmed control device, and can be used in multiple fields such as military, civil and commercial; the task information includes the take-off start and end point, the task type, the task deadline, the task content, the priority order and the drone allocation task priority table, wherein the task type includes two types: single task and collaborative task, the single task type refers to the task type that only needs to be completed by the current target drone, and the collaborative task type refers to the task type that needs to be completed by multiple drones including the current target drone; the task deadline refers to the latest completion time of the task; the take-off start and end point refers to the set take-off location and the set landing location of the target drone; the task content includes the performance constraint parameters of the target drone and specific task guidelines, such as shooting a specific area, tracking a moving target, collecting environmental data, etc.; wherein the performance constraint parameters include the maximum climb rate, the maximum turning angle, the flight altitude range, the maximum driving speed, the minimum driving speed, the maximum deceleration, etc.; the drone allocation task priority table refers to a list consisting of the numbers and priorities of all drones used for the current task when the task type is a collaborative task.

[0021] In this embodiment, the first route refers to a flight route obtained by analyzing and screening historical flight routes screened out based on mission information of the target UAV; the optimized route is a route obtained after sharp turn identification and smoothing of the first route.

[0022] The beneficial effects of the above technical solution are: by optimizing the first route determined according to the mission information of the target UAV, an optimized route is obtained; the flight status of the target UAV flying along the optimized route is monitored in real time, and when an abnormality is detected, the corresponding strategy is responded to for abnormal processing, which can effectively enhance the flight safety and flight efficiency of the UAV, thereby improving the execution efficiency of the mission.

[0023] The embodiment of the present invention provides a method for integrated management of UAV flight, which obtains and determines a first route according to mission information of a target UAV, including: Extracting a first starting point and a first end point from the mission information of the target UAV; Using the first starting point and the first end point as screening conditions, screening out historical flight routes that are consistent with the first starting point and the first end point from the drone record database; Determine the mission type based on the mission information of the target UAV; According to the mission type, the acquired historical flight routes are analyzed and screened to obtain the first route.

[0024] In this embodiment, the target UAV refers to the UAV that currently needs to perform a mission; the mission information includes the takeoff start and end points, mission type, mission deadline, mission content, priority order, and UAV task allocation priority table; the first starting point refers to the takeoff starting point in the mission information of the target UAV, that is, the set take-off location for the target UAV to perform the mission; the first end point refers to the takeoff end point in the mission information of the target UAV, that is, the set landing location for the target UAV to perform the mission.

[0025] In this embodiment, the drone record database is a database composed of historical flight route information of the drone, wherein the historical flight route information includes the drone number, takeoff start and end points, route trajectory, route passing points and flight time; the historical flight route refers to the historical flight route of the drone with the first starting point as the takeoff start point and the first end point as the landing point; the task type includes two types: single task and collaborative task, the single task type refers to the task type that only needs to be completed by the current target drone, and the collaborative task type refers to the task type that needs to be completed by multiple drones including the current target drone; the first route is the route obtained by analyzing and screening the acquired historical flight routes based on the task type.

[0026] The beneficial effect of the above technical solution is that by automatically screening historical flight routes according to the mission information of the target UAV, routes similar to the current mission can be quickly located, reducing the time and energy of planning the route from scratch and the risks during flight, which is conducive to improving the efficiency of mission execution.

[0027] The embodiment of the present invention provides a method for integrated management of unmanned aerial vehicle flight, which analyzes and screens the acquired historical flight routes according to the mission type to obtain a first route, including: When the mission type is a single mission type, if there is a single historical flight route, the current historical flight route is output as the first route; If there are multiple historical flight routes, the adaptation coefficient of each currently acquired historical flight route is calculated; The calculation formula of the adaptation coefficient is as follows: ; In the formula, Expressed as the adaptation coefficient of the i-th historical flight route; It is represented as the route length of the i-th historical flight route; It is expressed as the maximum flight speed of the current target UAV; It is represented by the influence weight of the estimated flight time of the UAV's flight route on the evaluation of route adaptation; It represents the maximum flight altitude of the current target UAV; It is represented as the maximum altitude of the i-th historical flight route; Expressed as the standard deviation of the altitude change of the i-th historical flight route; It is expressed as the weight of the impact of the altitude change of the flight route on the evaluation of route adaptation; It is expressed as the evaluation coefficient of avoiding restricted areas of the ith historical flight route; It is the weight of the ability to avoid restricted areas on the evaluation of route adaptation; Select the historical flight route with the highest adaptation coefficient as the first route output; If there is no historical flight route, a default route that matches the key matching parameters of the current target UAV is used as the first route output; When the mission type is a collaborative mission type, if there is a single historical flight route and the priority of the current target UAV is the first, the currently acquired historical flight route is regarded as the first route; If the priority of the current target UAV is not the first, a default route that matches the key matching parameters of the current target UAV is used as the first route output; If there is no historical flight route, a default route that matches the key matching parameters of the current target UAV is used as the first route output; If there are multiple historical flight routes, the historical flight route with an adaptation coefficient greater than the set adaptation threshold is regarded as the first candidate route; If there is a first candidate route, determining a first starting time for the current target UAV to travel the current first candidate route based on the priority order of the target UAVs; Based on the first starting time and the minimum driving speed of the target UAV, determine the first predicted time when the current target UAV completes the mission by traveling the current first candidate route; If the first predicted moments corresponding to all the first candidate routes obtained exceed the mission period of the current target UAV, a default route that matches the key matching parameters of the current target UAV is used as the first route output; If there is a first predicted time corresponding to the first candidate route that does not exceed the mission period of the current target UAV, the corresponding first candidate route is regarded as a candidate route; When there is only a single candidate route, the current candidate route is output as the first route; When there are multiple candidate routes, the candidate route with the highest fitness coefficient with the current target UAV is selected as the first route output.

[0028] In this embodiment, the task types include single tasks and collaborative tasks. The single task type refers to a task type that only needs to be completed by the current target drone, and the collaborative task type refers to a task type that needs to be completed by multiple drones including the current target drone.

[0029] In this embodiment, the adaptation coefficient is used to express the degree of adaptation between the current route and the target UAV; the evaluation coefficient for avoiding restricted areas refers to the ratio of the number of no-fly zones and restricted areas avoided by the current route to the total number of no-fly zones and restricted areas; the weight of the influence of the estimated flight time, altitude change and the ability to avoid restricted areas on the evaluation of route adaptation , as well as It is solved by using the matrix constructed after pairwise comparison and relative importance scoring using the hierarchical analysis method.

[0030] In this embodiment, the key matching parameters include a first starting point, a first end point, a flight altitude range, and a speed range, wherein the flight altitude range is composed of the maximum flight altitude and the minimum flight altitude of the target UAV, and the speed range is composed of the maximum flight speed and the minimum flight speed of the target UAV; the default route refers to a route obtained by route planning using the A* algorithm based on terrain data within a semicircle range with the flight starting point as the center and the straight-line distance from the flight starting point to the flight landing point as the radius, as well as different flight altitude ranges and flight speed ranges.

[0031] In this embodiment, for example, there is a target drone 1 with a first starting point of b1, a first end point of b2, and a flight speed range of , the flight altitude range is ; The default flight starting point of route 1 is b1, the flight landing point is b2, and the flight speed range is , the flight altitude range is ; The default flight starting point of route 2 is b1, the flight landing point is b2, and the flight speed range is , the flight altitude range is At this time, it is determined that the default route 2 is compatible with the target UAV 1, and the default route 2 is output as the first route.

[0032] In this embodiment, the adaptation threshold is set in advance and is generally 0.8; the first candidate route refers to a historical flight route with an adaptation coefficient greater than the set adaptation threshold among multiple historical flight routes when the mission type is a collaborative mission; the first starting time refers to the estimated start time of the current target UAV traveling on the first candidate route; the minimum driving speed refers to the minimum driving speed of the target UAV; the first predicted time refers to the predicted task completion time of the target UAV traveling on the first candidate route at the minimum driving speed; the mission deadline refers to the latest completion time for the target UAV to complete the assigned task; the candidate route refers to the corresponding first candidate route whose first predicted time does not exceed the mission deadline of the current target UAV.

[0033] The beneficial effect of the above technical solution is: by quickly selecting the historical flight route or default route with the highest adaptation coefficient based on the mission type, the time for the UAV to search and plan routes in the air is reduced, which is beneficial to improving mission execution efficiency, enhancing flight safety, improving mission flexibility and supporting collaborative tasks.

[0034] The embodiment of the present invention provides a method for integrated management of unmanned aerial vehicle flight, which optimizes the first route to obtain an optimized route, including: Traversing each sub-route in the first route to determine a first angle between adjacent sub-route; When there is a first angle greater than a set angle threshold, the corresponding sub-route is marked as a sharp turn segment; If there is only a single sharp turn segment, the current sharp turn segment is smoothed to obtain a new sub-route; Evaluating the new sub-route based on flight reliability to obtain a reliability evaluation coefficient; If the reliability evaluation coefficient is greater than the set evaluation threshold, the current route is determined to be reliable, and the corresponding sharp turn section is replaced with a new sub-route to generate an optimized route; If the reliability evaluation coefficient is not greater than the set evaluation threshold, the current route is determined to be unreliable, and the current sharp turn segment is smoothed again until a reliable new sub-route is obtained to replace the sharp turn segment and generate an optimized route; When there are multiple sharp turn segments, and the sharp turn segments are continuous, the first and last points of the continuous sharp turn segments are extracted and connected in a straight line to generate a first preset route; Evaluating the first preset route based on flight reliability to obtain a reliability evaluation coefficient; If the reliability evaluation coefficient is greater than the set evaluation threshold, the current route is determined to be reliable, and the first preset route is used to replace the corresponding continuous sharp turn segments to generate an optimized route; If the reliability evaluation coefficient is not greater than the set evaluation threshold, the current route is determined to be unreliable. At this time, the sharp turns with continuous relationships are smoothed in turn, and the reliable route replaces the corresponding sharp turns with continuous relationships, and finally generates an optimized route; A sharp turn segment that is not continuous with other sharp turn segments is marked as a first sharp turn segment and smoothed to obtain a second preset route; The effect of the second preset route is verified based on flight reliability. If the reliability evaluation coefficient is greater than a set evaluation threshold, the current route is determined to be reliable, and the corresponding first sharp turn section is replaced with the second preset route to generate an optimized route. If the reliability evaluation coefficient is not greater than the set evaluation threshold, the current route is determined to be unreliable, and the corresponding first sharp turn segment is smoothed again until a reliable second preset route is obtained to replace the first sharp turn segment, thereby generating an optimized route.

[0035] In this embodiment, a sub-route refers to a segmented route of a first route; a first angle refers to a turning angle between adjacent sub-routes; a set angle threshold refers to 80% of the maximum turning angle of the target UAV; a sharp turn segment refers to a sub-route whose first angle is greater than the set angle threshold; a new sub-route refers to a route obtained by smoothing the sharp turn segment using a terrain-based track smoothing method; a reliability evaluation coefficient is used to evaluate the reliability of the current route; and a set evaluation threshold is pre-set and is generally 0.7.

[0036] In this embodiment, the first preset route refers to a route obtained by extracting the first and last points of the continuous sharp turn segments and connecting them in a straight line when there are multiple sharp turn segments and the existing sharp turn segments are continuous; the first sharp turn segment refers to a sharp turn segment that is not continuous with other sharp turn segments; the second preset route is a route obtained by smoothing the first sharp turn segment using a terrain-based track smoothing method.

[0037] In this embodiment, for example, the first route 1 has sharp turn segments a1, a2, a3, a4 and a5 in sequence, among which the sharp turn segments a1, a2 and a3 are continuous, and the sharp turn segments a4 and a5 are not continuous with other sharp turn segments. At this time, the route obtained by connecting the first point of the sharp turn segment a1 and the last point of the sharp turn segment a3 in a straight line is used as the first preset route, and the sharp turn segments a4 and a5 are respectively marked as the first sharp turn segments.

[0038] The beneficial effect of the above technical solution is: by detecting and marking sharp turns and performing smoothing, the UAV can be effectively prevented from making sharp turns during flight, thereby reducing the risk of flight accidents that may be caused by sudden turns and ensuring the safety of UAV flight.

[0039] The embodiment of the present invention provides a method for integrated management of UAV flight, and the calculation formula of the reliability evaluation coefficient is as follows: ; In the formula, Y0 represents the reliability evaluation coefficient of the current route; It is expressed as the performance constraint satisfaction coefficient of the UAV on the current route; It is expressed as the weight of the impact of the performance constraint satisfaction capability on the reliability of the current route; It is expressed as the route length of the original sharp turn section; It is expressed as the route length of the current route; It is expressed as the weight of the impact of the change in route length on the reliability of the current route; It represents the set safety distance threshold between the flight route and the no-fly zone; It is represented by the distance between the jth no-fly zone and the current route that does not maintain a safe distance from the current route, where j=1, 2, 3, , n1; It is expressed as the total number of no-fly zones that do not maintain a safe distance from the current route; It indicates the set safety distance threshold between the route and the restricted area; It is represented by the distance between the kth no-fly zone and the current route that does not maintain a safe distance from the current route, where k=1, 2, 3, , n2; It is expressed as the total number of restricted areas that do not maintain a safe distance from the current route; It is expressed as the weight of the ability to avoid restricted areas on the reliability of the current route.

[0040] In this embodiment, the UAV performance constraint satisfaction coefficient refers to the ratio of the number of constraints that the current route satisfies the target UAV performance constraints to the total number of target UAV performance constraints, where the target UAV performance constraints are predetermined, including maximum climb rate, maximum turning angle, etc.; the weight of the impact of performance constraint satisfaction ability, route length change, and restricted area avoidance ability on the reliability of the current route , 2 and 3 is solved by using the matrix constructed after pairwise comparison and relative importance scoring using the analytic hierarchy process.

[0041] The beneficial effect of the above technical solution is that by calculating the reliability evaluation coefficient, a data basis can be provided for selecting a route to replace the sharp turn section, which is beneficial to reducing the risk during the flight of the UAV and reducing flight accidents that may be caused by sharp turns.

[0042] The embodiment of the present invention provides a method for integrated management of UAV flight, which monitors the flight status of the target UAV according to the optimized route in real time, and when an abnormality is detected, responds to a corresponding strategy to perform abnormality processing, including: Obtaining the initial battery power of the target UAV; When the initial battery power is greater than the set battery power, the target UAV flies according to the optimized route; When the initial battery power is not greater than the set battery power, the target drone sends a low-battery signal to the maintenance personnel; The maintenance personnel receive the low-battery signal and replace the battery of the current target drone; Using a set data collection tool to collect in real time the flight status of the target UAV flying along the optimized route to obtain first real-time flight data; Inputting the first real-time flight data into an anomaly detection model established in advance based on a machine learning algorithm to obtain a flight anomaly detection result; According to the flight anomaly detection result, if there is a flight anomaly type, sending an alarm signal to a ground control station; For abnormal situations where the flight abnormality type is the presence of dynamic obstacles, real-time dynamic obstacle avoidance operations are performed based on the real-time flight environment data of the target UAV; For anomalies where the flight anomaly type is not the presence of a dynamic obstacle, an anomaly handling strategy matching the current flight anomaly type is extracted from the set anomaly type-strategy mapping table to perform an anomaly handling operation; The exception handling process is recorded, a flight exception maintenance report is generated and stored in the flight exception record library.

[0043] In this embodiment, the initial battery power refers to the remaining battery power of the target UAV before the mission is performed; the set battery power refers to 80% of the total battery power of the target UAV; the set data collection tools include event cameras, RGB cameras and various sensors, such as GPS locators, altimeters, gyroscopes, accelerometers, etc.; the first real-time flight data includes position, speed, attitude, battery power, etc.; the anomaly detection model is a model obtained by training a neural network using the historical flight data of the UAV marked with flight anomaly types after preprocessing, which is used to identify the flight anomaly type of the current UAV; the flight anomaly detection result is obtained by inputting the first real-time data into the anomaly detection model; the flight anomaly types include the presence of dynamic obstacles, power anomaly, flight speed anomaly and deviation from the route; the set anomaly type-strategy mapping table is pre-established, consisting of the flight anomaly type and the corresponding anomaly handling strategy, for example, if the flight anomaly type is deviation from the route, the anomaly handling strategy is that the target UAV automatically adjusts the flight attitude and speed to return to the preset route; if the flight anomaly type is power anomaly, the anomaly handling strategy is that the target UAV immediately lands to the nearest emergency landing point; the flight anomaly maintenance report includes the time when the flight anomaly occurs, the flight anomaly type and the anomaly handling operation; the flight anomaly record library is used to store the flight anomaly maintenance report.

[0044] The beneficial effect of the above technical solution is: by real-time monitoring of the flight status of the target UAV flying according to the optimized route, and when an abnormality is detected, responding to the corresponding strategy to handle the abnormality, the flight safety and efficiency of the UAV can be improved.

[0045] The embodiment of the present invention provides a method for integrated management of unmanned aerial vehicle flight, which performs real-time dynamic obstacle avoidance operations based on real-time flight environment data of a target unmanned aerial vehicle for abnormal situations in which the flight abnormality type is the presence of dynamic obstacles, including: After filtering, denoising and motion compensation of the real-time flight environment data of the target UAV, a first event image is obtained; Performing dynamic target segmentation on the first event image to obtain a first detection result; Performing dynamic obstacle detection on the image collected by the RGB camera to obtain a second detection result; Fusing the first detection result and the second detection result to obtain a target detection result; Combining the target detection result with a depth camera, performing depth information analysis on the dynamic obstacle, and obtaining first depth information; The first depth information is used as an input quantity, input into a Kalman filter and a time delay is introduced, and a first position and a first speed of the current dynamic obstacle are output; Determining a first relative speed according to the first speed and the real-time speed of the current target UAV; Based on the first relative speed, determining a first speed obstacle area of ​​the current target UAV; Inputting the first position, the first speed, and the first speed obstacle area of ​​the current dynamic obstacle into the strategy generation model constructed by the pre-trained neural network, and outputting a first obstacle avoidance strategy; The target UAV executes the first obstacle avoidance strategy and performs dynamic obstacle avoidance operations.

[0046] In this embodiment, real-time flight environment data refers to an event stream collected by an event camera; the first event image is an event image generated by visualizing events in the form of a double threshold of event number and fixed time for the real-time flight environment data after filtering out background noise and hot spot noise, and the image is obtained after motion compensation; the first detection result is a result obtained by dynamic obstacle detection and segmentation of the first event image, including confidence, detection box, etc.; the second detection result is a result obtained by dynamic obstacle detection of an image collected by an RGB camera, including confidence, detection box, etc.; the target detection result is a fusion of the first detection result and the second detection result, including the fused confidence and detection box, etc.

[0047] In this embodiment, the first depth information refers to the distance information between the dynamic obstacle and the target UAV obtained by the depth camera; the Kalman filter is an algorithm that uses the linear system state equation to optimally estimate the system state through system input and output observation data; the first relative speed refers to the relative speed between the UAV and the dynamic obstacle obtained by vector calculation based on the first speed of the dynamic obstacle and the real-time speed of the current target UAV.

[0048] In this embodiment, the first speed obstacle area refers to an area that the target UAV should avoid entering, which is an area obtained by superimposing a basic influence area determined by geometric calculation using the size of the dynamic obstacle expanded 3 times with the current dynamic obstacle as the center, and a relative speed influence area calculated according to the first relative speed and the set safety margin of the target UAV, wherein the relative speed influence area is first obtained by using the first relative speed and the maximum deceleration of the target UAV to decelerate the UAV from the current speed to zero. After the deceleration distance is added to the set safety margin to obtain the reserved distance, the circular range divided by the reserved distance as the radius with the current dynamic obstacle as the center is used as the relative speed influence area; the strategy generation model is a model obtained by training a neural network by annotating the collected flight environment data with obstacle type, position, speed and obstacle avoidance strategy as training data, and is used to output obstacle avoidance strategies, such as speed adjustment and heading change of the UAV; the first obstacle avoidance strategy is an obstacle avoidance strategy obtained by inputting the first position, first speed and first speed obstacle area of ​​the current dynamic obstacle into the strategy generation model.

[0049] The beneficial effect of the above technical solution is: by filtering, denoising and motion compensating the real-time flight environment data of the target UAV, and then realizing dynamic obstacle avoidance operation, it is beneficial to enhance the target detection capability, accurately estimate the dynamic obstacle information, and obtain the obstacle avoidance strategy to achieve effective obstacle avoidance, thereby improving flight safety.

[0050] The embodiment of the present invention provides a drone flight integrated management system, such as Figure 2 As shown, including: Flight route acquisition module: used to obtain and determine the first route based on the mission information of the target UAV; A flight route optimization module: used to optimize the first route to obtain an optimized route; UAV flight monitoring module: used to monitor the flight status of the target UAV according to the optimized route in real time, and when an abnormality is detected, respond to the corresponding strategy for abnormal processing.

[0051] The beneficial effects of the above technical solution are: by optimizing the first route determined according to the mission information of the target UAV, an optimized route is obtained; the flight status of the target UAV flying along the optimized route is monitored in real time, and the flight status of the target UAV flying along the optimized route is monitored in real time, and when an abnormality is detected, the corresponding strategy is responded to for abnormal processing, which can effectively enhance the flight safety and flight efficiency of the UAV, thereby improving the execution efficiency of the mission.

[0052] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A method for integrated management of unmanned aerial vehicle flight, characterized in that: include: Step 1: Obtain and determine the first route based on the mission information of the target UAV; Step 2: Optimize the first route to obtain an optimized route; Step 3: Monitor the flight status of the target UAV according to the optimized route in real time, and when an abnormality is detected, respond to the corresponding strategy to handle the abnormality.

2. The method for integrated management of UAV flight according to claim 1, characterized in that: Obtain and determine the first route based on the mission information of the target UAV, including: Extracting a first starting point and a first end point from the mission information of the target UAV; Using the first starting point and the first end point as screening conditions, screening out historical flight routes that are consistent with the first starting point and the first end point from the drone record database; Determine the mission type based on the mission information of the target UAV; According to the mission type, the acquired historical flight routes are analyzed and screened to obtain the first route.

3. The method for integrated management of UAV flight according to claim 2, characterized in that: According to the mission type, the historical flight routes are analyzed and screened to obtain the first route, including: When the mission type is a single mission type, if there is a single historical flight route, the current historical flight route is output as the first route; If there are multiple historical flight routes, the adaptation coefficient of each currently acquired historical flight route is calculated; The calculation formula of the adaptation coefficient is as follows: ; In the formula, Expressed as the adaptation coefficient of the i-th historical flight route; It is represented as the route length of the i-th historical flight route; It is expressed as the maximum flight speed of the current target UAV; It is represented by the influence weight of the estimated flight time of the UAV's flight route on the evaluation of route adaptation; It represents the maximum flight altitude of the current target UAV; It is represented as the maximum altitude of the i-th historical flight route; Expressed as the standard deviation of the altitude change of the i-th historical flight route; It is expressed as the weight of the impact of the altitude change of the flight route on the evaluation of route adaptation; It is expressed as the evaluation coefficient of avoiding restricted areas of the ith historical flight route; It is the weight of the ability to avoid restricted areas on the evaluation of route adaptation; Select the historical flight route with the highest adaptation coefficient as the first route output; If there is no historical flight route, a default route that matches the key matching parameters of the current target UAV is used as the first route output; When the mission type is a collaborative mission type, if there is a single historical flight route and the priority of the current target UAV is the first, the currently acquired historical flight route is regarded as the first route; If the priority of the current target UAV is not the first, a default route that matches the key matching parameters of the current target UAV is used as the first route output; If there is no historical flight route, a default route that matches the key matching parameters of the current target UAV is used as the first route output; If there are multiple historical flight routes, the historical flight route with an adaptation coefficient greater than the set adaptation threshold is regarded as the first candidate route; If there is a first candidate route, determining a first starting time for the current target UAV to travel the current first candidate route based on the priority order of the target UAVs; Based on the first starting time and the minimum driving speed of the target UAV, determine the first predicted time when the current target UAV completes the mission by traveling the current first candidate route; If the first predicted moments corresponding to all the first candidate routes obtained exceed the mission period of the current target UAV, a default route that matches the key matching parameters of the current target UAV is used as the first route output; If there is a first predicted time corresponding to the first candidate route that does not exceed the mission period of the current target UAV, the corresponding first candidate route is regarded as a candidate route; When there is only a single candidate route, the current candidate route is output as the first route; When there are multiple candidate routes, the candidate route with the highest fitness coefficient with the current target UAV is selected as the first route output.

4. The method for integrated management of UAV flight according to claim 3, characterized in that: Mission information includes takeoff start and end points, mission type, mission deadline, mission content, priority, and drone assignment priority table.

5. The method for integrated management of unmanned aerial vehicle flight according to claim 1, characterized in that: Optimizing the first route to obtain an optimized route includes: Traversing each sub-route in the first route to determine a first angle between adjacent sub-route; When there is a first angle greater than a set angle threshold, the corresponding sub-route is marked as a sharp turn segment; If there is only a single sharp turn segment, the current sharp turn segment is smoothed to obtain a new sub-route; Evaluating the new sub-route based on flight reliability to obtain a reliability evaluation coefficient; If the reliability evaluation coefficient is greater than the set evaluation threshold, the current route is determined to be reliable, and the corresponding sharp turn section is replaced with a new sub-route to generate an optimized route; If the reliability evaluation coefficient is not greater than the set evaluation threshold, the current route is determined to be unreliable, and the current sharp turn segment is smoothed again until a reliable new sub-route is obtained to replace the sharp turn segment and generate an optimized route; When there are multiple sharp turn segments, and the sharp turn segments are continuous, the first and last points of the continuous sharp turn segments are extracted and connected in a straight line to generate a first preset route; Evaluating the first preset route based on flight reliability to obtain a reliability evaluation coefficient; If the reliability evaluation coefficient is greater than the set evaluation threshold, the current route is determined to be reliable, and the first preset route is used to replace the corresponding continuous sharp turn segments to generate an optimized route; If the reliability evaluation coefficient is not greater than the set evaluation threshold, the current route is determined to be unreliable. At this time, the sharp turns with continuous relationships are smoothed in turn, and the reliable route replaces the corresponding sharp turns with continuous relationships, and finally generates an optimized route; A sharp turn segment that is not continuous with other sharp turn segments is marked as a first sharp turn segment and smoothed to obtain a second preset route; The effect of the second preset route is verified based on flight reliability. If the reliability evaluation coefficient is greater than a set evaluation threshold, the current route is determined to be reliable, and the corresponding first sharp turn section is replaced with the second preset route to generate an optimized route. If the reliability evaluation coefficient is not greater than the set evaluation threshold, the current route is determined to be unreliable, and the corresponding first sharp turn segment is smoothed again until a reliable second preset route is obtained to replace the first sharp turn segment, thereby generating an optimized route.

6. The method for integrated management of UAV flight according to claim 5, characterized in that: The calculation formula of the reliability evaluation coefficient is as follows: ; In the formula, Y0 represents the reliability evaluation coefficient of the current route; It is expressed as the performance constraint satisfaction coefficient of the UAV on the current route; It is expressed as the weight of the impact of the performance constraint satisfaction capability on the reliability of the current route; It is expressed as the route length of the original sharp turn section; It is expressed as the route length of the current route; It is expressed as the weight of the impact of the change in route length on the reliability of the current route; It represents the set safety distance threshold between the flight route and the no-fly zone; It is represented by the distance between the jth no-fly zone and the current route that does not maintain a safe distance from the current route, where j=1, 2, 3, , n1; It is expressed as the total number of no-fly zones that do not maintain a safe distance from the current route; It indicates the set safety distance threshold between the route and the restricted area; It is represented by the distance between the kth no-fly zone and the current route that does not maintain a safe distance from the current route, where k=1, 2, 3, , n2; It is expressed as the total number of restricted areas that do not maintain a safe distance from the current route; It is expressed as the weight of the ability to avoid restricted areas on the reliability of the current route.

7. The integrated management method for UAV flight according to claim 1, characterized in that: Monitor the flight status of the target UAV according to the optimized route in real time, and when an abnormality is detected, respond to the corresponding strategy to handle the abnormality, including: Obtaining the initial battery power of the target UAV; When the initial battery power is greater than the set battery power, the target UAV flies according to the optimized route; When the initial battery power is not greater than the set battery power, the target drone sends a low-battery signal to the maintenance personnel; The maintenance personnel receive the low-battery signal and replace the battery of the current target drone; Using a set data collection tool to collect in real time the flight status of the target UAV flying along the optimized route to obtain first real-time flight data; Inputting the first real-time flight data into an anomaly detection model established in advance based on a machine learning algorithm to obtain a flight anomaly detection result; According to the flight anomaly detection result, if there is a flight anomaly type, sending an alarm signal to a ground control station; For abnormal situations where the flight abnormality type is the presence of dynamic obstacles, real-time dynamic obstacle avoidance operations are performed based on the real-time flight environment data of the target UAV; For anomalies where the flight anomaly type is not the presence of a dynamic obstacle, an anomaly handling strategy matching the current flight anomaly type is extracted from the set anomaly type-strategy mapping table to perform an anomaly handling operation; The exception handling process is recorded, a flight exception maintenance report is generated and stored in the flight exception record library.

8. The integrated management method for UAV flight according to claim 6, characterized in that: For abnormal situations where the flight abnormality type is the presence of dynamic obstacles, real-time dynamic obstacle avoidance operations are performed based on the real-time flight environment data of the target UAV, including: After filtering, denoising and motion compensation of the real-time flight environment data of the target UAV, a first event image is obtained; Performing dynamic target segmentation on the first event image to obtain a first detection result; Performing dynamic obstacle detection on the image collected by the RGB camera to obtain a second detection result; Fusing the first detection result and the second detection result to obtain a target detection result; Combining the target detection result with a depth camera, performing depth information analysis on the dynamic obstacle, and obtaining first depth information; The first depth information is used as an input quantity, input into a Kalman filter and a time delay is introduced, and a first position and a first speed of the current dynamic obstacle are output; Determining a first relative speed according to the first speed and the real-time speed of the current target UAV; Based on the first relative speed, determining a first speed obstacle area of ​​the current target UAV; Inputting the first position, the first speed, and the first speed obstacle area of ​​the current dynamic obstacle into the strategy generation model constructed by the pre-trained neural network, and outputting a first obstacle avoidance strategy; The target UAV executes the first obstacle avoidance strategy and performs dynamic obstacle avoidance operations.

9. An integrated management system for unmanned aerial vehicle flight, characterized in that: include: Flight route acquisition module: used to obtain and determine the first route based on the mission information of the target UAV; A flight route optimization module: used to optimize the first route to obtain an optimized route; UAV flight monitoring module: used to monitor the flight status of the target UAV according to the optimized route in real time, and when an abnormality is detected, respond to the corresponding strategy for abnormal processing.

Images