Unmanned aerial vehicle dynamic route planning method and system, electronic device and storage medium

Abstract

The application relates to a UAV dynamic flight route planning method, system, electronic equipment and storage medium. The UAV dynamic flight route planning method comprises the following steps: acquiring an initial flight route; acquiring a real-time target in the process of flying from the starting point of the initial flight route to the ending point of the initial flight route; matching the real-time target with a preset task planning template; generating a real-time target flight route according to the matching result, and splicing the real-time target flight route with the initial flight route to form a dynamic flight route; repeating the steps of acquiring the real-time target to forming the dynamic flight route until the ending point of the initial flight route is reached. In the application, the UAV dynamic flight route planning method introduces intelligent algorithms and route planning algorithms to build the response capability of the UAV to various types of sudden conditions, reduce the dependence on people in the process, and improve the intelligent degree of the UAV operation.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) technology, specifically to a UAV dynamic flight path planning method, system, electronic device, and storage medium. Background Technology

[0002] Unmanned aerial vehicles (UAVs) typically acquire current situational information about targets during their flight path by carrying sensors such as electro-optical pods, cameras, and lidar. UAVs offer advantages such as fast response times and unobstructed viewing angles, enabling timely acquisition of information about observed objects. However, traditional UAVs lack dynamic and intelligent flight capabilities. They generally only possess static data acquisition methods, making it difficult to meet the demands for real-time, dynamic data acquisition. For example, current technologies typically employ general flight mission planning methods, manually drawing point, line, and area targets and manually setting flight path planning parameters. By processing the mission blocks, a flight path is obtained, and the UAV flies and executes the mission according to this path. However, UAVs can fly at speeds up to 20 m / s; by the time human decision-making and planning occur after target detection, the UAV has already flown too far and missed the target.

[0003] Therefore, how to enable drones to react quickly and automatically on the airborne end and automatically plan mission routes is a key issue that urgently needs to be solved in the current development of intelligent and automated drones. Summary of the Invention

[0004] The technical problem to be solved by this invention is how to enable UAVs to react quickly and automatically on the airborne end and automatically plan mission routes.

[0005] One of the technical solutions adopted by this invention to solve its technical problem is: a dynamic flight path planning method for unmanned aerial vehicles (UAVs), comprising the following steps:

[0006] Obtain the initial flight path;

[0007] During the flight from the starting point of the initial flight path to the ending point of the initial flight path, real-time targets are acquired;

[0008] Match real-time goals with preset task planning templates;

[0009] The real-time target flight path is generated based on the matching results, and the real-time target flight path is spliced ​​with the initial flight path to form a dynamic flight path;

[0010] Repeat the steps of acquiring the real-time target to forming the dynamic flight path until the end of the initial flight path is reached.

[0011] Furthermore, obtaining the initial flight path includes the following steps:

[0012] Obtain the basic data for initial route construction based on task requirements;

[0013] The initial flight path is generated based on the initial route construction data.

[0014] Furthermore, the basic data includes initial mission objective data, environmental terrain data, flight performance parameter data, and payload performance constraint data.

[0015] Furthermore, the acquisition of the real-time target includes the following steps:

[0016] Acquire remote sensing observation data;

[0017] The remote sensing data is analyzed to obtain real-time targets;

[0018] Obtain the real-time geographic coordinates of the target.

[0019] Furthermore, the remote sensing observation data includes at least one data frame acquired through at least one sensing unit.

[0020] Furthermore, the step of parsing the remote sensing observation data to obtain the real-time target includes the following steps:

[0021] The remote sensing observation data is analyzed using target detection and recognition algorithms or segmentation algorithms;

[0022] The analysis results are compared with preset key targets to obtain real-time targets and identify the target type of the real-time targets.

[0023] Furthermore, it also includes the following steps:

[0024] Obtain real-time target constraint data;

[0025] Obtain the degree of correlation between constraint data among real-time targets in multiple different data frames;

[0026] Set a threshold for the degree of correlation;

[0027] If the correlation between the constraint data of the real-time targets of the multiple different data frames is greater than or equal to the threshold, then the real-time targets of the multiple different data frames are combined into the same real-time target.

[0028] Furthermore, matching the real-time target with the preset task planning template includes the following steps:

[0029] Obtain a vector data model, wherein the vector data model includes features corresponding to at least one real-time target;

[0030] The characteristics of the real-time target are analyzed based on the vector data model.

[0031] Based on the feature analysis results, a preset task planning template is matched to the real-time target.

[0032] Furthermore, the step of performing characteristic analysis on the real-time target based on the vector data model includes the following steps: describing the real-time target using a unified vector data model;

[0033] Map real-time targets to features of the vector data model;

[0034] The description information of the real-time target is stored as attribute data.

[0035] Furthermore, the step of matching a preset task planning template to the real-time target based on the feature analysis results includes the following steps:

[0036] The feature analysis results include a vector data model corresponding to the real-time target;

[0037] The vector data model corresponding to the real-time target is decomposed into a single real-time target and / or a set of real-time targets, wherein the set of real-time targets includes at least two real-time targets that are correlated with each other in the constraint data;

[0038] Preset task planning templates are matched for individual real-time targets and / or sets of real-time targets respectively.

[0039] Furthermore, the step of generating a real-time target flight path based on the matching results and then concatenating the real-time target flight path with the initial flight path to form a dynamic flight path includes the following steps:

[0040] Obtain the analysis of the machine body and load constraints;

[0041] The optimal path is planned based on the analysis results that conform to the constraints of the machine body and load.

[0042] Based on the optimal path planning results, multiple individual real-time targets and / or sets of real-time targets are matched with preset task planning templates and connected to form real-time target flight paths.

[0043] The real-time target flight path is combined with the initial flight path to form a dynamic flight path.

[0044] Another technical solution adopted by the present invention to solve its technical problem is: a dynamic flight path planning system for unmanned aerial vehicles, comprising:

[0045] The control terminal is used to form the initial flight path;

[0046] The drone terminal is used to execute the initial flight path formed by the control terminal;

[0047] A communication and control module is installed on the UAV and is used to acquire real-time targets.

[0048] An airborne data processing module is installed on the UAV. The airborne data processing module is used to match the real-time target with a preset mission planning template, generate a real-time target flight path based on the matching result, and splice the real-time target flight path with the initial flight path to form a dynamic flight path.

[0049] Furthermore, the sensing and control module includes:

[0050] A communication unit, which is used for communication between the UAV and the outside world;

[0051] A sensing unit is used to acquire real-time targets;

[0052] A control unit, which is used to control the UAV to perform flight missions.

[0053] Another technical solution adopted by the present invention to solve its technical problem is: an electronic device, comprising:

[0054] Memory, used to store computer programs;

[0055] A processor is used to execute the computer program to implement the steps of the UAV dynamic route planning method described above.

[0056] Another technical solution adopted by the present invention to solve its technical problem is: a computer-readable storage medium storing a computer program, wherein the computer program, when executed by a processor, implements the steps of the UAV dynamic route planning method described above.

[0057] As can be seen from the above technical solutions, the UAV dynamic flight path planning method in this application introduces intelligent algorithms and flight path planning algorithms to build the UAV's ability to cope with various types of emergencies, reduce reliance on human intervention, and improve the intelligence level of UAV operations. This enables the UAV to deviate from its preset flight path based on the dynamic evolution of current information when facing real-time targets, track key targets, and obtain effective observation information, thereby improving the effectiveness of task analysis and judgment. Secondly, the large amount of observation data collected using the UAV dynamic flight path planning method eliminates the need for manual analysis. After identifying and monitoring real-time targets, the UAV can directly match the corresponding task planning template based on the characteristics of the real-time targets and form a dynamic flight path, significantly improving the timeliness of observing and acquiring relevant information about real-time targets.

[0058] In this application, the UAV dynamic flight path planning system can realize the dynamic identification, decision-making, planning and operation of targets on the UAV terminal, simplify the information flow path and improve the timeliness of data collection.

[0059] Furthermore, the electronic devices and computer-readable storage media in this application can also achieve the above-mentioned effects. Attached Figure Description

[0060] The present invention will be further described below with reference to the accompanying drawings and embodiments;

[0061] Figure 1 This is a flowchart of the UAV dynamic route planning method in a specific embodiment of the present invention;

[0062] Figure 2 This is a flowchart illustrating the process of obtaining the initial flight path in a specific embodiment of the present invention;

[0063] Figure 3 This is a flowchart illustrating the acquisition of real-time targets in a specific embodiment of the present invention;

[0064] Figure 4 This is a flowchart illustrating the process of parsing the remote sensing observation data to obtain real-time targets in a specific embodiment of the present invention;

[0065] Figure 5 This is a flowchart illustrating how real-time targets are matched with preset task planning templates in a specific embodiment of the present invention;

[0066] Figure 6 This is a flowchart illustrating the characteristic analysis of the real-time target based on the vector data model in a specific embodiment of the present invention;

[0067] Figure 7 This is a flowchart illustrating the matching of a preset task planning template to a real-time target based on the characteristic analysis results in a specific embodiment of the present invention.

[0068] Figure 8 This is a flowchart illustrating the formation of a dynamic flight path in a specific embodiment of the present invention;

[0069] Figure 9 This is a flowchart illustrating the execution of the UAV dynamic flight path planning method by the UAV dynamic flight path planning system in a specific embodiment of the present invention;

[0070] Figure 10 This is a schematic diagram illustrating the construction method of dynamic flight routes in a specific embodiment of the present invention. Detailed Implementation

[0071] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments.

[0072] The technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the embodiments described are only one of the embodiments of the present invention, and not all of the embodiments. Other embodiments promoted by those skilled in the art based on the embodiments of the present invention are all within the scope of protection of the present invention.

[0073] This application provides the following embodiment, please refer to... Figure 1 As shown, a dynamic flight path planning method for unmanned aerial vehicles (UAVs) includes the following steps:

[0074] S1. Obtain the initial flight path;

[0075] S2. During the flight from the starting point of the initial flight path to the ending point of the initial flight path, acquire real-time targets;

[0076] S3. Match real-time targets with preset task planning templates;

[0077] S4. Generate a real-time target flight path based on the matching results, and then combine the real-time target flight path with the initial flight path to form a dynamic flight path;

[0078] S5. Repeat the steps of obtaining the real-time target to forming the dynamic flight path (i.e., repeat steps S2 to S4) until the end of the initial flight path is reached.

[0079] The aforementioned UAV dynamic flight path planning method can be a task-oriented replanning business process built on the UAV end. This means that during flight along the initial flight path, the UAV end can automatically detect and identify key real-time targets, plan real-time target flight paths based on the targets and their characteristics, and control the UAV end to execute these flight paths. Secondly, the real-time target flight paths, which currently require manual control, are ported to the airborne end using a task planning template combined with a general flight task planning algorithm. The UAV end then automatically executes the planning of real-time target flight paths, replacing human intervention. Furthermore, the real-time targets acquired by different sensing units on the UAV end are uniformly abstracted into a vector data model. Finally, while the general task planning template is universal, it is difficult for the UAV end to meet the specific operational requirements of different real-time targets when using a universal task planning template for real-time target flight path planning. Therefore, by using preset rules or parameters, real-time target flight paths can be planned to meet the task requirements of different targets in different scenarios.

[0080] Among them, real-time target matching with mission planning template refers to selecting the best mission planning template style that matches the real-time target from the mission planning template library according to the mission target characteristics and payload characteristics of the real-time target, instantiating each form item in the mission planning template, determining the sub-targets, time and space constraints of each real-time target unit, and generating the real-time target flight path.

[0081] In some specific embodiments, please refer to Figure 2 As shown, obtaining the initial flight path in step S1 includes the following steps:

[0082] S11. Obtain basic data for initial route construction based on task requirements; wherein, the task requirement data includes task objective data and observation requirement data;

[0083] S12. Generate an initial flight route based on the initial route construction basic data.

[0084] In some specific embodiments, the basic data includes initial mission objective data, environmental terrain data, flight performance parameter data, and payload performance constraint data.

[0085] The initial mission objective data includes information such as vector feature type, target type, and geographical location. Furthermore, the initial mission objective can also be to specify a path for observation or collection of specific information; this application does not impose specific limitations on this. Environmental terrain data includes digital elevation models of the area where the initial mission objective is located and the area corresponding to the current location of the UAV, as well as parameters such as wind direction and wind speed in the area corresponding to the current location of the UAV. Flight performance parameter data includes UAV constraint data, namely minimum turning radius, maximum flight distance, maximum rate of climb, maximum range, and remaining battery power. Payload performance constraint data includes the optical plane geometry model of the sensing unit and pre-calibrated persistent parameters.

[0086] In some specific embodiments, please refer to Figure 3 As shown, obtaining the real-time target in step S2 includes the following steps:

[0087] S21. Acquire remote sensing observation data;

[0088] S22. Analyze the remote sensing observation data to obtain real-time targets;

[0089] S23. Obtain the real geographic coordinates of the real-time target.

[0090] The real geographic coordinates of the real-time target can be obtained by acquiring the initial coordinates of the real-time target on the corresponding remote sensing observation data, and converting the initial coordinates into real geographic coordinates based on multiple relevant parameter data at the observation time and direct geographic positioning based on a rigorous imaging model.

[0091] For example, the data frames acquired using a camera as a sensing unit are remote sensing observation data. The coordinates of the real-time target in the coordinate system of the video frame image are the initial coordinates. Multiple relevant parameter data at the observation time include the UAV pose, sensing unit state parameters, sensing unit installation parameters, etc. By combining the above initial coordinates with the relevant parameter data at the observation time and combining the direct geolocation of the rigorous imaging model, the initial coordinates are converted into real geographic coordinates in the geodetic coordinate system.

[0092] In some specific embodiments, the remote sensing observation data includes at least one data frame acquired through at least one sensing unit.

[0093] In some specific embodiments, please refer to Figure 4 As shown, step S22, which involves parsing the remote sensing observation data to obtain the real-time target, includes the following steps:

[0094] S221. The remote sensing observation data is parsed using a target detection and recognition algorithm or a segmentation algorithm;

[0095] S222. Compare the analysis results with preset key targets to obtain real-time targets and identify the target type of the real-time targets.

[0096] Among them, target detection and recognition algorithms include, but are not limited to, network model extraction algorithms based on architectures such as convolutional neural networks (CNN) and Transformers.

[0097] In some specific embodiments, please refer to Figure 3 As shown, step S2 further includes the following steps:

[0098] S24. Obtain the constraint data of the real-time target;

[0099] S25. Obtain the degree of correlation between constraint data among real-time targets in multiple different data frames;

[0100] S26. Set a threshold for the degree of association;

[0101] S27. If the correlation degree of the constraint data between the real-time targets of the multiple different data frames is greater than or equal to the threshold, then the real-time targets of the multiple different data frames are combined into the same real-time target.

[0102] The constraint data for real-time targets includes the target type, spatial location, and observation time. The correlation degree can be set directly using a single constraint data point or using a function constructed from several constraints. For example, using the spatial location of the real-time target as the basis for setting the correlation degree, the reciprocal of the distance between the spatial locations of different real-time targets can be used as the correlation degree. In this case, the threshold can be the reciprocal of the accuracy of direct geolocation of the target. In other words, the closer two different real-time targets are, the larger their reciprocal distance will be. When the reciprocal of this distance is greater than the reciprocal of the progress of direct geolocation, the two real-time targets can be associated as a unified real-time target on the data frame, and the vector data model can be updated. This effectively reduces data redundancy, facilitates data calculation, and improves computational efficiency.

[0103] In some specific embodiments, please refer to Figure 5 As shown, the step S3 of matching the real-time target with the preset task planning template includes the following steps:

[0104] S31. Obtain a vector data model, wherein the vector data model includes at least one feature corresponding to a real-time target;

[0105] S32. Perform characteristic analysis on the real-time target based on the vector data model;

[0106] S33. Match the preset task planning template to the real-time target based on the feature analysis results.

[0107] Vector data models are data models used to represent and store spatial geometric data. They represent objects in space (such as points, lines, and polygons) as vectors with coordinates and attributes. Each object in a vector data model is represented by a set of coordinate values ​​that define the object's position and shape in space (i.e., the characteristics corresponding to the real-time target). Furthermore, vector data models can include additional attribute information, such as the object's name, type, and color. Vector data models can accurately represent and process the topological and geometric relationships between spatial objects. They support various spatial analysis operations, such as querying, buffer analysis, and overlay analysis. Vector data models can also perform spatial indexing and spatial query optimization, improving the efficiency of data retrieval and processing. Common vector data formats include Shapefile, GeoJSON, and KML. These formats can be used in Geographic Information System (GIS) software to create, edit, analyze, and visualize geographic data.

[0108] In some specific embodiments, please refer to Figure 6As shown, step S32, which involves analyzing the characteristics of the real-time target based on the vector data model, includes the following steps:

[0109] S321. Describe real-time targets using a unified vector data model;

[0110] S322. Map the real-time target to the features of the vector data model;

[0111] S323. Store the description information of the real-time target as attribute data.

[0112] For all types of sensing, the extracted real-time targets are described using a unified vector data model of geometric objects such as points, lines, and surfaces. This model includes, but is not limited to, the abstract base class Geometry and its first-level subclasses Point, Curve, Surface, and GeometryCollection, as well as their second-level subclasses LineString and Polygon. By using this unified vector data model, the position, shape, and attribute information of real-time targets can be accurately represented. This helps to more accurately understand and describe the characteristics and states of real-time targets, reduces data redundancy, and facilitates computation. Specifically, the descriptive information of real-time targets may include the real-time target type and observation time.

[0113] In some specific embodiments, please refer to Figure 7 As shown, step S33, which involves matching a preset task planning template to the real-time target based on the feature analysis results, includes the following steps:

[0114] S331, The characteristic analysis results include a vector data model corresponding to the real-time target;

[0115] S332. Decompose the vector data model corresponding to the real-time target into a single real-time target and / or a set of real-time targets, wherein the set of real-time targets includes at least two real-time targets that are correlated with each other in the constraint data;

[0116] S333: Match preset task planning templates to a single real-time target and / or a set of real-time targets respectively.

[0117] In a specific embodiment, the vector data model is decomposed into several individual real-time targets, or it can be decomposed into several sets of real-time targets. Each set of real-time targets includes at least two real-time targets that are related to each other in the constraint data. For example, two real-time targets located at adjacent locations can be decomposed into the same set of real-time targets. The setting of the set of real-time targets can be such that there is a relationship between one set of constraint data and no relationship between other constraint data, or that there is a relationship between several sets of constraint data but no relationship between one set of constraint data. The constraint data includes the type of real-time target, the spatial location of the real-time target, the observation time, etc.

[0118] In some specific embodiments, please refer to Figure 8 As shown, step S4, which involves generating a real-time target flight path based on the matching result and then concatenating the real-time target flight path with the initial flight path to form a dynamic flight path, includes the following steps:

[0119] S41. Obtain the body and load constraint analysis;

[0120] S42. Perform optimal path planning based on the analysis results that conform to the aforementioned body and load constraints;

[0121] S43. Based on the optimal path planning results, multiple individual real-time targets and / or sets of real-time targets are matched with preset task planning templates and connected to form real-time target flight paths.

[0122] S44. The real-time target flight path is spliced ​​with the initial flight path to form a dynamic flight path.

[0123] The aforementioned airframe and load constraint analysis refers to the analysis and interpretation of the UAV's airframe structure and various loads (such as sensors, equipment, and payloads) to determine the limitations and constraints it faces during flight.

[0124] Airframe constraint analysis primarily focuses on the physical structure and mechanical performance of the UAV. Airframe constraints include evaluating and analyzing aspects such as the UAV's size, weight, materials, strength, and stiffness to ensure that the UAV's structure can withstand the forces and loads experienced during flight and operation when planning dynamic flight paths, while meeting the requirements of safety, stability, and reliability.

[0125] Load constraint analysis focuses on the characteristics and requirements of the various payloads carried by the UAV. It considers limitations on payload weight, size, power consumption, data transmission requirements, etc., to ensure that the UAV can effectively carry and operate these payloads when planning dynamic flight paths, and to meet the requirements of specific missions or applications.

[0126] Among them, optimal path planning refers to path planning with the goal of achieving the optimal efficiency in completing real-time target reconnaissance, under the premise of satisfying the constraints of the body and the load.

[0127] This application also provides an embodiment of a dynamic flight path planning system for unmanned aerial vehicles (UAVs), comprising:

[0128] The control terminal is used to form the initial flight path;

[0129] The drone terminal is used to execute the initial flight path formed by the control terminal;

[0130] A communication and control module is installed on the UAV and is used to acquire real-time targets.

[0131] An airborne data processing module is installed on the UAV. The airborne data processing module is used to match the real-time target with a preset mission planning template, generate a real-time target flight path based on the matching result, and splice the real-time target flight path with the initial flight path to form a dynamic flight path.

[0132] The control terminal can be a ground control station, a cloud control platform, or other control equipment. The control terminal includes a data communication unit. The control terminal can upload the mission flight path to the communication and control module of the UAV through the data communication unit. The data communication unit can transmit uplink control data from the positioning and attitude determination system sensors, as well as downlink GPS, flight status, and various airborne sensor status and observation data to the UAV.

[0133] In some specific embodiments, the sensing control module includes:

[0134] A communication unit, which is used for communication between the UAV and the outside world;

[0135] A sensing unit is used to acquire real-time targets;

[0136] A control unit, which is used to control the UAV to perform flight missions.

[0137] The control unit includes a flight controller and an application controller. The control unit can decompose the flight path into flight control and application control. The flight controller controls the attitude, altitude, and speed of the UAV based on information such as UAV navigation and inertial navigation. The navigation module supports multiple navigation and positioning systems such as GPS, GLONASS, and BeiDou, and provides RTK function to provide accurate navigation information. The application controller can operate various types of sensing units to meet the needs of different application scenarios. For example, it can trigger optical array sensors to perform fixed-distance, fixed-point, or timed exposures, and can provide AHRS information to the electro-optical pod and perform control operations such as searching, scanning, and tracking of the pod according to instructions. Sensing units include, but are not limited to, remote sensing payloads such as optical array sensors, electro-optical pods, and lidar.

[0138] The sensing unit can collect remote sensing observation data along the flight path under the guidance and control of the application controller.

[0139] The airborne data processing module is capable of real-time data processing, including preprocessing of remote sensing observation data, feature extraction of remote sensing observation data, and real-time detection of preset key targets through target detection and recognition algorithms or segmentation algorithms. It identifies target types and obtains the initial coordinates of real-time targets on raw observation data such as data frames or image data. Key targets refer to targets that need to be observed during the mission, such as power transmission lines, towers, wildfires, vehicles, machinery, landslides, etc. Key targets can be adjusted or set according to mission requirements. Based on the UAV pose, sensing unit status parameters, sensing unit installation parameters, etc. at the observation time, the initial coordinates of real-time targets are converted into real geographic coordinates based on direct geolocation using a rigorous imaging model.

[0140] In addition, the airborne data processing module can also dynamically generate real-time target flight paths on the UAV based on the vector data model of the real-time target, taking into account factors such as the target characteristics, geographical and meteorological environment, payload characteristics, and UAV constraints, with the goal of achieving the best efficiency in completing the real-time target reconnaissance. Based on the core portable flight path planning SDK deployed on the airborne end, it can also generate real-time target flight paths on the UAV end.

[0141] This application also provides an embodiment of a UAV dynamic flight path planning system for executing a UAV dynamic flight path planning method. The UAV dynamic flight path planning system includes a ground control station, a UAV terminal, an airborne data processing module, and a communication, sensing, and control module. The ground control station includes functions such as mission flight path planning, remote UAV control, and real-time mission monitoring. The UAV terminal is a carrier equipped with the communication, sensing, and control module and the airborne data processing module for observation tasks. The communication, sensing, and control module includes a communication unit, a sensing unit, and an intelligent control unit. The communication unit, sensing unit, and intelligent control unit are integrated and deployed as an integrated module on the UAV terminal, enabling remote real-time integrated communication, sensing, and control of the UAV terminal, reducing command data transfer time, achieving autonomous decision-making during flight, and acquiring effective observation data. The communication unit is responsible for communication between the UAV and the ground control station and / or between the UAV and the cloud control platform; the perception unit mainly refers to sensors, including but not limited to photoelectric pods, aerial survey cameras, and lidar, which are mainly used to acquire real-time target observation data; the control unit includes a motion controller and an application controller. The motion controller mainly controls the flight navigation of the UAV and guides the UAV to perform maneuvering control such as attitude, speed, and altitude. The application controller mainly guides the UAV to complete task-oriented data acquisition control, including the UAV's approach to targets, obstacle avoidance, and other actions, as well as the action control of the mission payload; the airborne data processing module is used for real-time processing of sensor data and dynamic route planning.

[0142] As described above, by constructing a drone-based task replanning workflow, the drone can automatically detect and identify key real-time targets while flying along the initial flight path. It then plans real-time target flight paths based on the targets and their characteristics and controls the drone to execute these paths. Secondly, the real-time target flight paths, which currently require manual control, are ported to the airborne terminal using a task planning template combined with a general flight task planning algorithm. The drone then automatically executes the planning of these flight paths, replacing human intervention. Furthermore, the real-time targets acquired by the different sensing units on the drone are uniformly abstracted into a vector data model. Finally, while the general task planning template is universal, it is difficult for the drone to meet the specific operational requirements of different real-time targets when using a universal template for real-time target flight path planning. Therefore, by using preset rules or parameters, real-time target flight paths can be planned to meet the task requirements of different targets in different scenarios.

[0143] In a specific embodiment, please refer to Figure 9 As shown, when the UAV dynamic flight path planning system executes the UAV dynamic flight path planning method, it can be done according to the following steps:

[0144] S01. The ground control station generates an initial flight path that meets the constraints of the airframe and payload based on mission requirements, mission objectives, environmental terrain, and other information, and uploads the initial flight path to the control unit on the UAV.

[0145] S02. The UAV flies under the guidance and control of the control unit, and the data communication unit can transmit control data and remote sensing observation data from the sensing unit.

[0146] S03. During the process of acquiring remote sensing observation data, the sensing unit uses target detection and recognition algorithms or segmentation algorithms to detect preset key real-time targets in real time, obtain the initial coordinates of the real-time targets on the original observation data such as data frames or image data of remote sensing observation data, and convert the initial coordinates of the real-time targets into real geographic coordinates based on the UAV pose, sensing unit status parameters, sensing unit installation parameters, etc. at the observation time.

[0147] S04. Based on factors such as real-time target characteristics, geographical and meteorological environment, payload characteristics, and UAV constraints, with the goal of achieving the optimal completion efficiency of real-time target reconnaissance, a task style template is matched, and the real-time target flight path is dynamically generated on the UAV through the airborne data processing module. The real-time target flight path is then spliced ​​with the initial flight path to form a dynamic flight path.

[0148] S05. The UAV repeats steps S03 to S04 until it reaches the end of the initial flight path.

[0149] The types of real-time targets can include: power transmission lines, power poles, wildfires, vehicles, machinery, landslides, etc. Different task planning templates are available for different real-time targets, including: power transmission line line inspection, detailed power pole inspection, continuous wildfire monitoring, vehicle tracking, and landslide survey. The task planning templates specifically include the UAV's flight path and payload actions for the real-time target. By instantiating the task planning templates, an ordered connection of the spatiotemporal nodes of the optimal flight path points and payload action points is constructed to form the real-time target flight path.

[0150] Taking wildfire hovering observation as an example, the identified wildfire hotspot is taken as the real-time target. Based on the target type of wildfire and the distance between the real-time target's geographical location and the current UAV position being less than a set threshold, the task planning template is matched as the "continuous wildfire observation" template. Therefore, hovering routes at different altitudes centered on the wildfire are constructed for the wildfire target. The route altitude and hovering radius are set according to the UAV's body constraints and target attributes. Based on the payload's own ground projection geometry model, parameters such as the mission payload attitude, field of view, and action commands are calculated. The optical array sensor is set to perform N fixed-point exposures on the hovering path to acquire multi-view image data, or the photoelectric pod tracking control operation is set to track the target center to acquire video data with continuously changing perspectives. Considering the constraints of the UAV's current position and remaining battery power, connecting multiple real-time target unit routes completes the generation of the real-time target flight route. The real-time target flight route is inserted into the UAV's current flight route. The current waypoint of the UAV's current flight route is connected to the first waypoint of the real-time target flight route, and the last waypoint of the real-time target flight route is connected to the next waypoint of the UAV's current flight route, ultimately forming a dynamic flight route.

[0151] The above-mentioned method for constructing dynamic flight routes can be referred to Figure 10 As shown.

[0152] The aforementioned UAV dynamic flight path planning method, by introducing intelligent algorithms and flight path planning algorithms, builds the UAV's ability to respond to various types of emergencies, reduces reliance on human intervention, and improves the intelligence level of UAV operations. It enables UAVs to deviate from preset flight paths based on the dynamic evolution of current information when facing real-time targets, tracking key targets and acquiring effective observation information, thereby improving the effectiveness of task analysis and judgment. Secondly, the large amount of observation data collected using the UAV dynamic flight path planning method eliminates the need for manual analysis. After identifying and monitoring real-time targets, the UAV can directly match the corresponding task planning template based on the characteristics of the real-time targets and form a dynamic flight path, significantly improving the timeliness of observing and acquiring relevant information about real-time targets. Simultaneously, the aforementioned UAV dynamic flight path planning system enables dynamic target identification, decision-making, planning, and operation at the UAV level, simplifying information flow paths and improving the timeliness of data collection.

[0153] This application also provides an embodiment of an electronic device, comprising:

[0154] Memory, used to store computer programs;

[0155] A processor is used to execute the computer program to implement the steps of the UAV dynamic route planning method described above.

[0156] This application also provides an embodiment of a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps of the UAV dynamic flight path planning method described above.

[0157] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media provided in this application and in the embodiments may include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual-speed SDRAM (SSRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).

[0158] It should be noted that, in this document, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, apparatus, article, or method that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, apparatus, article, or method. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, apparatus, article, or method that includes that element.

[0159] The foregoing has shown and described 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 to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of protection claimed by the present invention. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for dynamic flight path planning of unmanned aerial vehicles (UAVs), characterized in that, Includes the following steps: Obtain the initial flight path; During the flight from the starting point of the initial flight path to the ending point of the initial flight path, real-time targets are acquired; Match real-time goals with preset task planning templates; The real-time target flight path is generated based on the matching results, and the real-time target flight path is spliced ​​with the initial flight path to form a dynamic flight path; Repeat the steps of acquiring the real-time target to forming the dynamic flight path until the end of the initial flight path is reached; The step of generating a real-time target flight path based on the matching result and then combining the real-time target flight path with the initial flight path to form a dynamic flight path includes the following steps: Obtain the analysis of the machine body and load constraints; The optimal path is planned based on the analysis results that conform to the constraints of the machine body and load. Based on the optimal path planning results, multiple individual real-time targets and / or sets of real-time targets are matched with preset task planning templates and connected to form real-time target flight paths. The real-time target flight path is combined with the initial flight path to form a dynamic flight path.

2. The UAV dynamic flight path planning method according to claim 1, characterized in that, Obtaining the initial flight path includes the following steps: Obtain the basic data for initial route construction based on task requirements; The initial flight path is generated based on the initial route construction data.

3. The UAV dynamic flight path planning method according to claim 2, characterized in that, The basic data includes initial mission objective data, environmental terrain data, flight performance parameter data, and payload performance constraint data.

4. The UAV dynamic flight path planning method according to claim 1, characterized in that, The process of acquiring the real-time target includes the following steps: Acquire remote sensing observation data; The remote sensing data is analyzed to obtain real-time targets; Obtain the real-time geographic coordinates of the target.

5. The UAV dynamic flight path planning method according to claim 4, characterized in that, The remote sensing observation data includes at least one data frame acquired through at least one sensing unit.

6. The UAV dynamic flight path planning method according to claim 4, characterized in that, The process of parsing the remote sensing observation data to obtain the real-time target includes the following steps: The remote sensing observation data is analyzed using target detection and recognition algorithms or segmentation algorithms; The analysis results are compared with preset key targets to obtain real-time targets and identify the target type of the real-time targets.

7. The UAV dynamic flight path planning method according to claim 5, characterized in that, It also includes the following steps: Obtain real-time target constraint data; Obtain the degree of correlation between constraint data among real-time targets in multiple different data frames; Set a threshold for the degree of correlation; If the correlation between the constraint data of the real-time targets of the multiple different data frames is greater than or equal to the threshold, then the real-time targets of the multiple different data frames are combined into the same real-time target.

8. The UAV dynamic flight path planning method according to claim 1, characterized in that, The process of matching real-time targets with preset task planning templates includes the following steps: Obtain a vector data model, wherein the vector data model includes features corresponding to at least one real-time target; The characteristics of the real-time target are analyzed based on the vector data model. Based on the feature analysis results, a preset task planning template is matched to the real-time target.

9. The UAV dynamic flight path planning method according to claim 8, characterized in that, The step of performing characteristic analysis on the real-time target based on the vector data model includes the following steps: Real-time targets are described using a unified vector data model; Map real-time targets to features of the vector data model; The description information of the real-time target is stored as attribute data.

10. The UAV dynamic flight path planning method according to claim 9, characterized in that, The step of matching a preset task planning template to a real-time target based on the feature analysis results includes the following steps: The feature analysis results include a vector data model corresponding to the real-time target; The vector data model corresponding to the real-time target is decomposed into a single real-time target and / or a set of real-time targets, wherein the set of real-time targets includes at least two real-time targets that are correlated with each other in the constraint data; Preset task planning templates are matched for individual real-time targets and / or sets of real-time targets respectively.

11. A dynamic flight path planning system for unmanned aerial vehicles (UAVs), characterized in that, The method for performing the UAV dynamic flight path planning method as described in any one of claims 1-10 includes: The control terminal is used to form the initial flight path; The drone terminal is used to execute the initial flight path formed by the control terminal; A communication and control module is installed on the UAV and is used to acquire real-time targets. An airborne data processing module is installed on the UAV. The airborne data processing module is used to match the real-time target with a preset mission planning template, generate a real-time target flight path based on the matching result, and splice the real-time target flight path with the initial flight path to form a dynamic flight path.

12. The UAV dynamic flight path planning system according to claim 11, characterized in that, The sensing control module includes: A communication unit, which is used for communication between the UAV and the outside world; A sensing unit is used to acquire real-time targets; A control unit, which is used to control the UAV to perform flight missions.

13. An electronic device, characterized in that, include: Memory, used to store computer programs; A processor, configured to implement the steps of the UAV dynamic route planning method as described in any one of claims 1-10 when executing the computer program.

14. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program that, when executed by a processor, implements the steps of the UAV dynamic flight path planning method as described in any one of claims 1 to 10.

Images