A method, apparatus and related media for unmanned aerial vehicle (UAV) flight path planning

Abstract

This invention discloses a method, apparatus, and related media for drone flight path planning. The method includes constructing point cloud data from a set of project scene resources to obtain a point cloud target image data set; then configuring timeline tracks, setting safety constraints, and editing paths to generate an initial flight path data set; performing flight path calculation based on the initial flight path data set to obtain a target flight path data set; and finally, using the target flight path data set for debugging to obtain a debug flight path data set, which is then exported and deployed to obtain a control script data set. This invention utilizes the calculated target flight path data set for debugging in a 3D simulation environment, and then exports and deploys the debug flight path data set. Thus, drone flight path planning can satisfy both performance effects and flight feasibility.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) technology, and in particular to a method, apparatus, and related medium for UAV flight path planning. Background Technology

[0002] In the field of drone performance, existing animation path editing is mostly based on general 3D animation or timeline editing tools, which only focus on the generation of target patterns or visual effects. Such methods often separate the visual animation design from the drone's flight characteristics and safety constraints, and lack unified modeling and constraint transfer of constraints such as drone speed, acceleration and inter-drone safety distance. This results in a lack of fine constraint control in the conversion process from the target image to the executable flight path, making it difficult to detect and avoid potential trajectory intersections and collision risks in a timely manner. It is impossible to ensure that the generated multi-drone flight paths meet the performance effect while taking into account flight feasibility. Summary of the Invention

[0003] This invention provides a method, apparatus, and related medium for planning the flight path of unmanned aerial vehicles (UAVs), aiming to solve the technical problem that the flight path design of multiple UAVs in the prior art cannot simultaneously meet the performance effect and the flight feasibility.

[0004] In a first aspect, embodiments of the present invention provide a method for planning the flight path of an unmanned aerial vehicle (UAV), comprising:

[0005] Obtain the project scene resource set, and construct point cloud data from the project scene resource set to obtain the point cloud target image data set;

[0006] Configure the timeline track of the point cloud target image data set to obtain the timeline control data set;

[0007] Safety constraints are set using the time axis control data set to obtain the flight constraint data set;

[0008] The flight constraint data set is processed by path editing to generate an initial flight path data set in a custom timeline;

[0009] Based on the initial flight path data set, flight path settlement processing is performed to obtain the target flight path data set;

[0010] The target flight path data set is used to perform debugging processing in a three-dimensional simulation environment to obtain a debug flight path data set;

[0011] The debug flight path data set is exported and deployed to obtain the control script data set.

[0012] Secondly, embodiments of the present invention provide a drone flight path planning device, comprising:

[0013] The data acquisition unit is used to acquire a set of project scene resources and construct point cloud data from the set of project scene resources to obtain a set of point cloud target image data.

[0014] A time control unit is used to configure the time axis track of the point cloud target image data set to obtain a time axis control data set.

[0015] A safety constraint unit is used to set safety constraints using the time axis control data set to obtain a flight constraint data set.

[0016] The path editing unit is used to perform path editing processing on the flight constraint data set to generate an initial flight path data set in a custom timeline;

[0017] The path settlement unit is used to perform flight path settlement processing based on the initial flight path data set to obtain the target flight path data set.

[0018] The data debugging unit is used to perform debugging processing on the target flight path data set in a three-dimensional simulation environment to obtain a debugged flight path data set.

[0019] The path output unit is used to export and deploy the debug flight path data set to obtain a control script data set.

[0020] Thirdly, embodiments of the present invention provide a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the unmanned aerial vehicle (UAV) flight path planning method of the first aspect.

[0021] Fourthly, embodiments of the present invention provide a computer-readable storage medium, wherein a computer program is stored on the computer-readable storage medium, and when the computer program is executed by a processor, it implements the UAV flight path planning method of the first aspect.

[0022] This invention provides a method for planning the flight path of a drone, including acquiring a set of project scene resources and constructing point cloud data from the project scene resource set to obtain a point cloud target image data set; configuring the point cloud target image data set on a time axis to obtain a time axis control data set; setting safety constraints using the time axis control data set to obtain a flight constraint data set; performing path editing processing on the flight constraint data set to generate an initial flight path data set in a custom time axis; performing flight path calculation processing based on the initial flight path data set to obtain a target flight path data set; performing debugging processing on the target flight path data set in a 3D simulation environment to obtain a debug flight path data set; and exporting and deploying the debug flight path data set to obtain a control script data set. This invention, by performing debugging processing on the calculated target flight path data set in a 3D simulation environment and then exporting and deploying the debug flight path data set, allows the planning of drone flight paths to satisfy both performance effects and flight feasibility.

[0023] This invention also provides a drone flight path planning device, a computer device, and a storage medium, which have the same beneficial effects as described above. Attached Figure Description

[0024] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 A flowchart illustrating a method for planning flight paths of an unmanned aerial vehicle (UAV) provided in an embodiment of the present invention;

[0026] Figure 2 This is a schematic block diagram of a drone flight path planning device provided in an embodiment of the present invention. Detailed Implementation

[0027] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0028] It should be understood that, when used in this specification and the appended claims, the terms "comprising" and "including" indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0029] It should also be understood that the terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. As used in this specification and the appended claims, the singular forms “a,” “an,” and “the” are intended to include the plural forms unless the context clearly indicates otherwise.

[0030] It should also be further understood that the term "and / or" as used in this specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, and includes such combinations.

[0031] Please see below. Figure 1 , Figure 1 The flowchart of a UAV flight path planning method provided in an embodiment of the present invention specifically includes steps S101 to S107.

[0032] S101. Obtain the project scene resource set, and construct point cloud data from the project scene resource set to obtain the point cloud target image data set.

[0033] S102. Configure the time axis track of the point cloud target image data set to obtain the time axis control data set;

[0034] S103. Use the time axis control data set to set safety constraints and obtain a flight constraint data set;

[0035] S104. Perform path editing processing on the flight constraint data set to generate an initial flight path data set in a custom timeline;

[0036] S105. Perform flight path calculation processing based on the initial flight path data set to obtain the target flight path data set;

[0037] S106. The target flight path data set is used to perform debugging processing in a three-dimensional simulation environment to obtain a debug flight path data set;

[0038] S107. Export and deploy the debug flight path data set to obtain the control script data set.

[0039] In step S101, a set of project scene resources related to the drone performance project is obtained. The set of project scene resources includes various types of performance design resources such as images, models and animations used to describe the performance scenes. The above resources are uniformly analyzed and geometrically abstracted in a 3D editing environment, and the performance patterns and dynamic scenes are converted into a set of point cloud target scene data used to describe the target formation of drone formations.

[0040] In one embodiment, step S101 includes:

[0041] The project scene resource set is obtained by using the 3D engine editing environment to create and import resources.

[0042] Image resources are filtered from the project scene resource set to extract performance design diagrams and obtain an image data set;

[0043] Based on the image data set, extract the outline lines of the performance scene frame by frame to obtain the image outline line data set;

[0044] The image outline data set is processed to generate point coordinates, and a first point cloud data set is generated within a preset pixel interval range.

[0045] The 3D animation model data associated with the first point cloud data set is parsed to extract the vertex coordinates of each key frame, thus obtaining the animation vertex coordinate data set;

[0046] The animation vertex coordinate data set is processed by point cloud mapping to convert the keyframe vertices into drone points, thus obtaining a second point cloud data set;

[0047] The first point cloud data set and the second point cloud data set are merged to construct a point cloud target image data set.

[0048] In this embodiment, a drone performance project is created based on a 3D engine editing environment. For example, a new project is created in the Unity3D engine, and the image models, 3D animations, and related materials required for the drone performance design are uniformly imported through the resource import interface, so that the above image resources and 3D model animation resources form a project scene resource set in the project. After the import is completed, the project scene resource set is identified and filtered by resource type, and the performance design images used to describe the performance pattern are selected to form an image data set. The image format can include common vector or bitmap formats such as PNG, JPG, and SVG. For the image data set, a point cloud conversion tool (such as the SwarmPainter tool developed in this application) can be used to perform grayscale preprocessing on each performance design image. Under a unified coordinate system, the image contour is analyzed frame by frame and edge detection is performed according to a preset pixel density to extract the contour line information in the image, resulting in an image contour line data set. Based on this, discrete sampling points are generated along the contour lines according to the equidistant sampling strategy. The sampling interval can be configured within a preset pixel interval range. For example, the sampling interval is adjustable between 5 and 100 pixels. By default, sampling points are generated at a density of 50 pixels / point, thereby forming the first point cloud data set used to describe the contour of the static performance pattern.

[0049] Furthermore, for the 3D animation model resources associated with the first point cloud data set, it supports importing modeling and animation files in OBJ, FBX, and other formats exported from 3D modeling / animation software (such as Maya, 3ds Max, Blender, etc.), and parsing the aforementioned 3D animation models using a point cloud conversion tool. Specifically, firstly, the timeline and keyframe information in the animation file are parsed to locate the keyframes corresponding to each performance action; then, the vertex data of the model is traversed on each keyframe to extract the coordinate positions of the vertices in 3D space, constructing an animation vertex coordinate data set, so that each keyframe corresponds to a set of 3D vertices reflecting the structure of the performance scene. Based on this, point cloud mapping processing is performed on the animation vertex coordinate data set, transforming and normalizing the vertex coordinates of each keyframe according to the actual flight coordinate system usable by the drone formation, so that each vertex position can directly correspond to the target point position of the drone, thereby obtaining a second point cloud data set used to describe the dynamic performance scene. Finally, the first point cloud data set obtained from the static performance design images and the second point cloud data set obtained from the 3D animation model are merged and organized in a unified coordinate system. According to the performance script, the point cloud images of different patterns and different time periods are marked and classified to obtain the point cloud target image data set for subsequent timeline track configuration and path planning.

[0050] In step S102, the point cloud target image data set is loaded into the custom timeline editing interface. In the timeline, corresponding orbit layers are configured for elements such as UAV position, lighting and sound effects, and the control relationship of point cloud image evolution over time is established to generate a timeline control data set for driving subsequent flight path planning and calculation.

[0051] In one embodiment, step S102 includes:

[0052] The timeline component is loaded and an editing interface is constructed based on the point cloud target image data set to generate the timeline interface data set;

[0053] Configure the time precision of the time axis interface data set to obtain the configured time axis interface data set;

[0054] Configure the UAV position trajectory according to the configured timeline interface data set to obtain the UAV position trajectory data set;

[0055] The location track data set is used to configure the light track, so as to bind the light switch and color parameters to the track event, and obtain the UAV light track data set;

[0056] Based on the light track data set, configure the sound effect track and parse the audio to load the audio onto the track and extract the beat points to obtain the time axis control data set.

[0057] In this embodiment, a custom timeline editing component is created in a 3D engine editing environment (e.g., based on the Unity3D engine). The obtained point cloud target image data set is loaded into this timeline component, and a basic timeline editing interface is built on this basis. The editing interface includes a timeline scale display area and multiple track layers for carrying different data types, thereby obtaining a timeline interface data set for describing the timeline's visual layout and track configuration status. The time precision of the timeline interface data set is then configured, setting the minimum time scale of the timeline to a predetermined precision value, such as 10 milliseconds, to ensure fine-grained time alignment and synchronization control when adding events or keyframes to various tracks subsequently, resulting in a configured timeline interface data set.

[0058] After configuring the time accuracy, the drone position tracks are configured based on the configuration timeline interface data set. A corresponding data structure is defined for each position track, enabling it to accommodate multiple drone objects. The spatial coordinate information of each drone at different time points is recorded in the time dimension of the track. Specifically, this includes coordinate values ​​(x, y, z) describing the drone's three-dimensional position and association identifiers with point cloud target image data. This yields a drone position track data set used to control the spatial pose evolution of drone formations. The drone position track data set is then used to further configure lighting tracks. A separate track layer is defined for the drone's lighting effects in the timeline, and a track data structure is established for the lighting tracks, including fields such as light on / off status, color parameters (e.g., represented in RGB numerical form), and axis lock status. These lighting parameters are bound to corresponding time slices or keyframe events in the timeline, thereby generating a drone lighting track data set used to control the drone's lighting changes over time.

[0059] Furthermore, a sound track is configured in the timeline. Audio files used as background music for the performance are imported into the sound track. By parsing the audio signal, feature information such as beat points and accent positions in the audio is extracted, and these beat points are mapped to event nodes on the timeline. This allows the sound track to work collaboratively with the position track and lighting track on the same time reference. As a result, a timeline control data set integrating multiple control information such as drone position track, lighting track, and sound track is obtained, which is used to drive subsequent path planning and performance effect control.

[0060] In one embodiment, step S102 further includes:

[0061] The time axis control data set is initialized with time slices, and editable time slice objects are generated on each track to obtain the basic time slice data set.

[0062] Drag monitoring is performed on the time slice data set to generate time slices based on the drag position and length of the track and associate them with the performance segments to obtain the action time slice data set;

[0063] Perform time editing on the action time slice data set to obtain an optimized action time slice data set;

[0064] The interaction logic is configured using the optimized action time slice data set to define parameter editing, flight light control, and cross-orbit copy operations, thereby obtaining the interaction time slice set;

[0065] Based on the interactive time slice set, when the time slice is moved or adjusted, the key frames are aligned according to the preset key frame snapping rules to obtain an optimized time axis control data set.

[0066] In this embodiment, after configuring the timeline tracks and obtaining the timeline control data set, time slice initialization processing is first performed on various tracks, including drone position tracks, lighting tracks, and sound effect tracks, based on this timeline control data set. Editable time slice objects are generated in the timeline area of ​​each track, and the time slice serves as the basic unit for carrying performance script segments, thus obtaining the basic time slice data set. Users can create time slices by dragging the mouse on the tracks in the editing interface. The start and end positions of the drag and the current time scale of the track are automatically calculated to determine the start time and duration of the time slice. Each time slice is then associated with a corresponding performance action segment, which can be a predefined performance segment such as pattern formation, pattern transformation, or formation movement. Simultaneously, time slices can be stretched by dragging the boundaries to adjust their duration, and a time slice can be divided into multiple sub-time slices or adjacent time slices can be merged, thereby completing the drag monitoring processing of the time slice data set and the construction of the action time slice data set.

[0067] Based on this, the action time slice data set is processed by time editing. By limiting the start and end times of time slices to prevent illegal overlap and finely aligning time slices according to the performance rhythm, the time distribution of time slices on each track is optimized, resulting in an optimized action time slice data set. This optimized action time slice data set is then used to configure time slice-level interactive logic: when a user double-clicks a time slice in the editing interface, a parameter editing panel pops up. In this panel, the drone's flight parameters (such as flight speed, acceleration, etc.) and lighting parameters (such as light on / off status, color parameters) can be set for the corresponding time segment. It also supports copying and pasting time slice content, dragging and synchronously copying time slices between different tracks, allowing the same action segment to be quickly mapped to multiple tracks, resulting in an interactive time slice set. To ensure accurate synchronization between time slices and keyframes, a keyframe snapping rule can be introduced based on the interactive time slice set. When the user moves or adjusts the boundary of a time slice, if the edge of the time slice is close to a preset keyframe time point on the track, the edge of the time slice will automatically snap to the keyframe position within a time deviation range of less than a threshold. This avoids small but cumulative time misalignment between the time slice and the keyframe. Finally, after completing the above keyframe snapping and time alignment processing, an optimized time axis control data set is updated to guide subsequent constraint settings and path planning.

[0068] In step S103, based on the time axis control data set and flight parameters such as UAV model, body size, maximum flight speed, and maximum acceleration, the inter-machine safety distance, flight altitude range, and obstacle avoidance conditions are set within the planned space. Constraint modeling is performed on the point cloud target images corresponding to each time period on the time axis to obtain a flight constraint data set used to constrain the actual flight behavior of the UAV.

[0069] In step S104, the flight constraint data set is processed by path editing in the custom timeline interface, the UAV formation is mapped to each target screen point, and the key time points and intermediate transition paths are interactively adjusted according to the temporal relationship of the performance script. Under the premise of meeting the flight constraints, an initial flight path data set describing the trajectory of each UAV over time is generated.

[0070] In one embodiment, step S104 includes:

[0071] The flight constraint data set is initialized with a custom timeline to group the drones, resulting in a drone group data set;

[0072] A storyboard interface is constructed on the aforementioned UAV grouped data set to obtain a storyboard image identifier data set;

[0073] The point cloud is bound using the storyboard image identifier data set to obtain the storyboard point cloud image data set;

[0074] Configure the timeline of the storyboard point cloud image data set to obtain storyboard-bound flight constraint data;

[0075] Based on the flight constraint data bound to the storyboard, time-point visualization adjustments are made to generate an initial flight path data set.

[0076] In this embodiment, a custom timeline is initialized based on the generated flight constraint data set within the aforementioned timeline environment. The drones participating in the performance are grouped according to performance requirements, such as by screen area, formation type, or task role. Independent track identifiers are created for each drone group in the editing interface, resulting in a drone grouping data set describing the relationship between each group of drones and the constraints. The storyboard management panel is enabled in the editing interface, and multiple storyboard units are created according to the performance screen area or script logic, such as "Storyboard 1 - Moon," "Storyboard 2 - Beauty," etc. Each storyboard is then bound to the point cloud image data generated from the original performance design data, resulting in a storyboard image identifier data set that uniquely identifies the point cloud image corresponding to each storyboard.

[0077] After completing the storyboard labeling, the point cloud images corresponding to each storyboard are dragged and dropped onto a custom timeline in the form of time slices, and placed on the corresponding orbital positions of the target drone groups. This binds the storyboard point cloud images to specific drone groups, ensuring that the point cloud images presented by each drone group within each time period correspond one-to-one, thus obtaining a set of storyboard point cloud image data. Then, according to the preset sequence and duration of the performance script, the timeline configuration of the storyboard point cloud image data set is performed. The start and end times, transitions, and associations with flight constraints for each storyboard are determined on the timeline, generating storyboard-bound flight constraint data that integrates the storyboard images with flight constraints. Editors can preview and adjust this configuration through visual operations on the timeline: when playing the timeline, the corresponding storyboard point cloud images loaded by the drones in different time periods are displayed in a 3D view according to the time progress. Editors can observe the scene transitions and formation evolution, and make fine adjustments to key time points by dragging time slice boundaries and moving storyboard positions. After the above-mentioned visualization preview and time point adjustment, the flight constraint data of each segment is automatically integrated under the premise of meeting the established flight constraints, and an initial flight path data set is generated for subsequent path settlement and simulation debugging.

[0078] In step S105, based on the initial flight path data set, and combined with the flight characteristics of the UAV such as speed, acceleration and turning ability, the displacement process between different scenes is calculated, and the path coordinates, motion time and dynamic process are uniformly checked and optimized, and the target flight path data set that meets the performance and constraint conditions is output.

[0079] In one embodiment, step S105 includes:

[0080] The initial flight path data image point cloud is subjected to coordinate unification processing to obtain an aligned point cloud image data set;

[0081] The alignment point cloud image data set is used to perform matching calculations to calculate the spatial transformation parameters of image alignment and generate transformation parameter data;

[0082] Path planning is performed on the transformation parameter data to obtain the corresponding displacement path data;

[0083] Based on the drone's speed and acceleration, the displacement path data is processed by screen switching calculation to generate animation path data;

[0084] The animation path data is used for transition settlement to generate a transition frame point cloud split along the timeline, resulting in a target flight path data set.

[0085] In this embodiment, point cloud image data corresponding to adjacent performance image segments are extracted from the obtained initial flight path data. For example, on the timeline, the previous image is "Image 1 (moon pattern)" and the next image is "Image 2 (high-rise building pattern)". The point cloud data of these two image segments on the timeline are read and their coordinates are transformed under a unified three-dimensional reference system. All point clouds from different sources are mapped to the coordinate system actually used by the UAV during flight, thereby completing the coordinate unification processing of the point cloud image data and obtaining an aligned point cloud image data set. Based on the aligned two sets of point clouds, image transformation matching and alignment calculations are performed. The spatial transformation relationship between the two sets of point clouds is solved through geometric operations to obtain spatial transformation parameters describing the overall transformation between Image 1 and Image 2. The spatial transformation parameters include at least translation, rotation, and scaling components and are stored as transformation parameter data in the form of a matrix or equivalent data structure.

[0086] After obtaining the transformation parameter data, each drone's position in frame 1 is used as the starting point, and its corresponding target position in frame 2 is used as the ending point. Based on the aforementioned spatial transformation relationship, path planning is performed for each drone trajectory. Specifically, the initial displacement direction and displacement distance for each drone are first determined according to the transformation parameters. Then, the path type is selected based on this. For example, a straight path is used in areas with no obstructions and minimal changes, while a smooth curve path is used in areas where obstacles need to be avoided or where visual smoothness needs to be enhanced. At the same time, flight constraints are introduced during the planning process, including the drone's maximum flight speed, maximum acceleration, and safe distance between drones. The turning radius and rate of change of speed of the path are constrained and optimized to avoid generating motion trajectories that exceed the drone's dynamic performance or pose a collision risk, thereby obtaining displacement path data that meets flight characteristics and safety constraints. The above displacement path is solved for scene switching on the time dimension of the time axis: According to the preset duration of scene switching on the time axis, the transformation process from scene one to scene two is divided into several time sampling points. At each sampling moment, interpolation calculation is performed on all drones along the corresponding displacement path to obtain the set of drone spatial positions at that moment. Thus, the animation path data describing the scene switching process from "moon" to "tall building" is obtained.

[0087] Finally, the obtained animation path data is used for further transition calculation. The entire scene transformation process is broken down into continuous transition frames according to the timeline scale. For each frame, the corresponding drone point cloud layout is calculated, generating a time-series ordered set of transition frame point clouds. This ensures continuity and smoothness in spatial location and temporal distribution between adjacent frames. By combining all transition frame point clouds in chronological order, a complete target flight path data set can be constructed. This dataset simultaneously plots the position evolution trajectory of each drone throughout the performance and the dynamic switching relationships between different scene segments, providing a precise path foundation for subsequent 3D simulation debugging and control script generation.

[0088] In step S106, the target flight path data set is loaded into the three-dimensional simulation environment. The virtual UAV formation is used to simulate flight according to the target flight path. The formation changes, screen switching and overall appearance at different time periods are previewed in real time. During the simulation, parameters such as the timing and spatial position of key path nodes and transition sections and viewing angles are adjusted and corrected to obtain a set of debug flight path data for actual execution.

[0089] In one embodiment, step S106 includes:

[0090] The target flight path data set is loaded into a 3D simulation environment for real-time preview processing to obtain a preview result data set.

[0091] Based on the preview result data set, adjust the key point positions and transition parameters to obtain the debug flight path data set;

[0092] Based on the set of debug flight path data, the screen segments that need to be adjusted independently and their corresponding time ranges are determined, and a set of candidate screen fine-tuning data is obtained.

[0093] The viewing angle is simulated to adjust the display parameters of the candidate data set for fine-tuning the image, resulting in an optimized data set for image display.

[0094] The script organizes the optimized data set of the screen display to generate a set of debug flight path data.

[0095] In this embodiment, the target flight path data set obtained from the settlement is loaded into a simulation scene built based on a 3D engine. For example, a virtual scene corresponding to the actual performance venue is built in the Unity3D environment. The flight paths of each drone are mapped to the motion trajectories of virtual drone objects. The real-time preview function provided by the 3D engine is called to play the entire performance process in a timeline manner, obtaining a preview result data set for recording the visual effects, drone trajectory changes, and lighting coordination at different time points. Operators can play, pause, and view frame by frame in the simulation interface from different time points to intuitively observe whether the screen transitions are smooth, whether the formation changes are smooth, and whether there are any visual abruptness.

[0096] After obtaining the preview result data set, the parameters of key path nodes and transition segments are adjusted and optimized based on the trajectory and visual performance reflected within. If it is found that the drone's transitions are too abrupt, the rhythm of visual changes is inconsistent, or the lighting changes do not match the music beat within a certain time period, the corresponding key points can be located on the timeline. The spatial position, timestamp, and interpolation method between these key points and preceding and following key points can be adjusted, or the duration and interpolation curve of the visual transition segment can be modified, thereby generating an updated set of debug flight path data. Based on this, the overall performance script can be further divided according to the debug flight path data set, identifying visual segments requiring individual fine-tuning and their corresponding time ranges, such as a close-up shot or a transition segment between themes. These visual segments are marked as candidate data sets for visual fine-tuning, for subsequent targeted optimization. For the candidate data set for fine-tuning the visuals, a viewing angle simulation function is provided in the 3D simulation environment. By adjusting parameters such as the position, angle of view, and focal length of the virtual camera, the viewing effect of the audience in different viewing positions (such as the front grandstand, the side area, or the distant area) is simulated. Editors manually review the display effect of each candidate image under various viewing angles, focusing on whether the image composition is reasonable, whether the theme pattern is clear, and whether the image size and proportion are appropriate. They also adjust the image angle, image scaling ratio, and view cropping range accordingly in the simulation interface, thereby obtaining the optimized visual display data set. Finally, based on the optimized visual display data set, the performance script is organized and updated. The flight path and image configuration, after multiple rounds of preview debugging, key point correction, and viewing angle optimization, are solidified into a debug flight path data set that can be exported later. This provides a fully tested and optimized path data foundation for generating the final control script.

[0097] In step S107, the debug flight path data set is converted into a control script data set that conforms to the communication protocol and execution format requirements of the UAV control system, and the control script data set is deployed to the UAV ground control system or upper control equipment so that the UAV formation can execute the flight according to the planned path in the actual performance.

[0098] In summary, this application, based on custom timeline editing and animated path planning technology within a 3D engine environment, organically integrates point cloud image modeling, fine timeline control, multi-drone collaborative choreography, and flight characteristics and safety constraints. Compared to the traditional method of manually adjusting trajectories using general animation tools, it significantly improves editing efficiency in the design phase of drone performance paths and animations, enabling creators to quickly complete path planning and script generation for complex performance ideas on a unified platform. Through fine control of multiple tracks such as position, lighting, and sound effects on the timeline, and the integrated setting of multi-drone formation behavior, it helps ensure the overall performance... The sequential continuity and spatial coordination of the performance enhance the overall visual effect of the drone formation show. Simultaneously, by introducing flight constraints such as maximum speed, acceleration, and safe distance between drones during path calculation, the transitions between scenes and trajectories are checked and optimized. This effectively reduces collision risks while meeting the expected visual presentation, enhancing the safety of the performance. Furthermore, the dynamic path planning and collaborative control approach based on point cloud representation and time axis driving developed in this invention has certain promotional and reference value for similar multi-subject motion control scenarios such as robot formation performances and virtual scene animation production.

[0099] Combination Figure 2 As shown, Figure 2 This is a schematic block diagram of a drone flight path planning device provided in an embodiment of the present invention. The drone flight path planning device 200 includes:

[0100] The data acquisition unit 201 is used to acquire a project scene resource set and construct point cloud data from the project scene resource set to obtain a point cloud target image data set.

[0101] Time control unit 202 is used to configure the time axis track of the point cloud target image data set to obtain a time axis control data set;

[0102] Safety constraint unit 203 is used to set safety constraints using the time axis control data set to obtain flight constraint data set;

[0103] The path editing unit 204 is used to perform path editing processing on the flight constraint data set to generate an initial flight path data set in a custom timeline;

[0104] The path settlement unit 205 is used to perform flight path settlement processing based on the initial flight path data set to obtain the target flight path data set.

[0105] Data debugging unit 206 is used to perform debugging processing on the target flight path data set in a three-dimensional simulation environment to obtain a debugged flight path data set;

[0106] The path output unit 207 is used to export and deploy the debug flight path data set to obtain a control script data set.

[0107] In this embodiment, the data acquisition unit 201 acquires a project scene resource set and constructs point cloud data from the project scene resource set to obtain a point cloud target image data set; the time control unit 202 configures the time axis track of the point cloud target image data set to obtain a time axis control data set; the safety constraint unit 203 sets safety constraints using the time axis control data set to obtain a flight constraint data set; the path editing unit 204 performs path editing processing on the flight constraint data set to generate an initial flight path data set in a custom time axis; the path settlement unit 205 performs flight path settlement processing based on the initial flight path data set to obtain a target flight path data set; the data debugging unit 206 performs debugging processing using the target flight path data set in a three-dimensional simulation environment to obtain a debug flight path data set; and the path output unit 207 exports and deploys the debug flight path data set to obtain a control script data set.

[0108] In one embodiment, the data acquisition unit 201 is specifically used for:

[0109] The project scene resource set is obtained by using the 3D engine editing environment to create and import resources.

[0110] Image resources are filtered from the project scene resource set to extract performance design diagrams and obtain an image data set;

[0111] Based on the image data set, extract the outline lines of the performance scene frame by frame to obtain the image outline line data set;

[0112] The image outline data set is processed to generate point coordinates, and a first point cloud data set is generated within a preset pixel interval range.

[0113] The 3D animation model data associated with the first point cloud data set is parsed to extract the vertex coordinates of each key frame, thus obtaining the animation vertex coordinate data set;

[0114] The animation vertex coordinate data set is processed by point cloud mapping to convert the keyframe vertices into drone points, thus obtaining a second point cloud data set;

[0115] The first point cloud data set and the second point cloud data set are merged to construct a point cloud target image data set.

[0116] In one embodiment, the time control unit 202 is specifically used for:

[0117] The timeline component is loaded and an editing interface is constructed based on the point cloud target image data set to generate the timeline interface data set;

[0118] Configure the time precision of the time axis interface data set to obtain the configured time axis interface data set;

[0119] Configure the UAV position trajectory according to the configured timeline interface data set to obtain the UAV position trajectory data set;

[0120] The location track data set is used to configure the light track, so as to bind the light switch and color parameters to the track event, and obtain the UAV light track data set;

[0121] Based on the light track data set, configure the sound effect track and parse the audio to load the audio onto the track and extract the beat points to obtain the time axis control data set.

[0122] In one embodiment, the time control unit 202 is further specifically used for:

[0123] The time axis control data set is initialized with time slices, and editable time slice objects are generated on each track to obtain the basic time slice data set.

[0124] Drag monitoring is performed on the time slice data set to generate time slices based on the drag position and length of the track and associate them with the performance segments to obtain the action time slice data set;

[0125] Perform time editing on the action time slice data set to obtain an optimized action time slice data set;

[0126] The interaction logic is configured using the optimized action time slice data set to define parameter editing, flight light control, and cross-orbit copy operations, thereby obtaining the interaction time slice set;

[0127] Based on the interactive time slice set, when the time slice is moved or adjusted, the key frames are aligned according to the preset key frame snapping rules to obtain an optimized time axis control data set.

[0128] In one embodiment, the path editing unit 204 is specifically used for:

[0129] The flight constraint data set is initialized with a custom timeline to group the drones, resulting in a drone group data set;

[0130] A storyboard interface is constructed on the aforementioned UAV grouped data set to obtain a storyboard image identifier data set;

[0131] The point cloud is bound using the storyboard image identifier data set to obtain the storyboard point cloud image data set;

[0132] Configure the timeline of the storyboard point cloud image data set to obtain storyboard-bound flight constraint data;

[0133] Based on the flight constraint data bound to the storyboard, time-point visualization adjustments are made to generate an initial flight path data set.

[0134] In one embodiment, the path settlement unit 205 is specifically used for:

[0135] The initial flight path data image point cloud is subjected to coordinate unification processing to obtain an aligned point cloud image data set;

[0136] The alignment point cloud image data set is used to perform matching calculations to calculate the spatial transformation parameters of image alignment and generate transformation parameter data;

[0137] Path planning is performed on the transformation parameter data to obtain the corresponding displacement path data;

[0138] Based on the drone's speed and acceleration, the displacement path data is processed by screen switching calculation to generate animation path data;

[0139] The animation path data is used for transition settlement to generate a transition frame point cloud split along the timeline, resulting in a target flight path data set.

[0140] In one embodiment, the data debugging unit 206 is specifically used for:

[0141] The target flight path data set is loaded into a 3D simulation environment for real-time preview processing to obtain a preview result data set.

[0142] Based on the preview result data set, adjust the key point positions and transition parameters to obtain the debug flight path data set;

[0143] Based on the set of debug flight path data, the screen segments that need to be adjusted independently and their corresponding time ranges are determined, and a set of candidate screen fine-tuning data is obtained.

[0144] The viewing angle is simulated to adjust the display parameters of the candidate data set for fine-tuning the image, resulting in an optimized data set for image display.

[0145] The script organizes the optimized data set of the screen display to generate a set of debug flight path data.

[0146] Since the embodiments of the apparatus and the embodiments of the method correspond to each other, please refer to the description of the embodiments of the method for the embodiments of the apparatus, which will not be repeated here.

[0147] This invention also provides a computer-readable storage medium storing a computer program thereon, which, when executed, can perform the steps provided in the above embodiments. The storage medium may include various media capable of storing program code, such as a USB flash drive, a portable hard drive, a read-only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0148] This invention also provides a computer device, which may include a memory and a processor. The memory stores a computer program, and when the processor calls the computer program in the memory, it can implement the steps provided in the above embodiments. Of course, the computer device may also include various network interfaces, a power supply, a graphics card, etc., to utilize the graphics card's performance to operate the model, such as for inference and training.

[0149] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the systems disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple; relevant parts can be referred to in the method section. It should be noted that those skilled in the art can make various improvements and modifications to this application without departing from the principles of this application, and these improvements and modifications also fall within the protection scope of the claims of this application.

[0150] It should also be noted that, in this specification, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

Claims

1. A method for planning flight paths of unmanned aerial vehicles (UAVs), characterized in that, include: Obtain the project scene resource set, and construct point cloud data from the project scene resource set to obtain the point cloud target image data set; Configure the timeline track of the point cloud target image data set to obtain the timeline control data set; Safety constraints are set using the time axis control data set to obtain the flight constraint data set; The flight constraint data set is processed by path editing to generate an initial flight path data set in a custom timeline; Based on the initial flight path data set, flight path settlement processing is performed to obtain the target flight path data set; The target flight path data set is used to perform debugging processing in a three-dimensional simulation environment to obtain a debug flight path data set; The debug flight path data set is exported and deployed to obtain the control script data set; The step of configuring the timeline track of the point cloud target image data set to obtain a timeline control data set includes: loading a timeline component and constructing an editing interface based on the point cloud target image data set to generate a timeline interface data set; configuring the time precision of the timeline interface data set to obtain a configured timeline interface data set; configuring the UAV position track based on the configured timeline interface data set to obtain a UAV position track data set; configuring a light track using the position track data set to bind light on / off and color parameters to track events to obtain a UAV light track data set; and configuring a sound effect track based on the light track data set and parsing the audio to load the audio onto the track and extract beat points to obtain a timeline control data set. The step of performing path editing processing on the flight constraint data set to generate an initial flight path data set in a custom timeline includes: initializing the flight constraint data set with a custom timeline to group the drones, obtaining a drone group data set; constructing a storyboard interface on the drone group data set to obtain a storyboard image identifier data set; binding point clouds using the storyboard image identifier data set to obtain a storyboard point cloud image data set; configuring the timeline on the storyboard point cloud image data set to obtain storyboard-bound flight constraint data; and performing time point visualization adjustments based on the storyboard-bound flight constraint data to generate the initial flight path data set.

2. The UAV flight path planning method according to claim 1, characterized in that, The process of acquiring a project scene resource set and constructing point cloud data from the project scene resource set to obtain a point cloud target image data set includes: The project scene resource set is obtained by using the 3D engine editing environment to create and import resources. Image resources are filtered from the project scene resource set to extract performance design diagrams and obtain an image data set; Based on the image data set, extract the outline lines of the performance scene frame by frame to obtain the image outline line data set; The image outline data set is processed to generate point coordinates, and a first point cloud data set is generated within a preset pixel interval range. The 3D animation model data associated with the first point cloud data set is parsed to extract the vertex coordinates of each key frame, thus obtaining the animation vertex coordinate data set; The animation vertex coordinate data set is processed by point cloud mapping to convert the keyframe vertices into drone points, thus obtaining a second point cloud data set; The first point cloud data set and the second point cloud data set are merged to construct a point cloud target image data set.

3. The UAV flight path planning method according to claim 1, characterized in that, The step of configuring the timeline track of the point cloud target image data set to obtain the timeline control data set also includes: The time axis control data set is initialized with time slices, and editable time slice objects are generated on each track to obtain the basic time slice data set. Drag monitoring is performed on the time slice data set to generate time slices based on the drag position and length of the track and associate them with the performance segments to obtain the action time slice data set; Perform time editing on the action time slice data set to obtain an optimized action time slice data set; The interaction logic is configured using the optimized action time slice data set to define parameter editing, flight light control, and cross-orbit copy operations, thereby obtaining the interaction time slice set; Based on the interactive time slice set, when the time slice is moved or adjusted, the key frames are aligned according to the preset key frame snapping rules to obtain an optimized time axis control data set.

4. The UAV flight path planning method according to claim 1, characterized in that, The step of performing flight path calculation processing based on the initial flight path data set to obtain the target flight path data set includes: The initial flight path data image point cloud is subjected to coordinate unification processing to obtain an aligned point cloud image data set; The alignment point cloud image data set is used to perform matching calculations to calculate the spatial transformation parameters of image alignment and generate transformation parameter data; Path planning is performed on the transformation parameter data to obtain the corresponding displacement path data; Based on the drone's speed and acceleration, the displacement path data is processed by screen switching calculation to generate animation path data; The animation path data is used for transition settlement to generate a transition frame point cloud split along the timeline, resulting in a target flight path data set.

5. The UAV flight path planning method according to claim 1, characterized in that, The process of using the target flight path data set in a three-dimensional simulation environment to perform debugging processing to obtain a debug flight path data set includes: The target flight path data set is loaded into a 3D simulation environment for real-time preview processing to obtain a preview result data set. Based on the preview result data set, adjust the key point positions and transition parameters to obtain the debug flight path data set; Based on the set of debug flight path data, the screen segments that need to be adjusted independently and their corresponding time ranges are determined, and a set of candidate screen fine-tuning data is obtained. The viewing angle is simulated to adjust the display parameters of the candidate data set for fine-tuning the image, resulting in an optimized data set for image display. The script organizes the optimized data set of the screen display to generate a set of debug flight path data.

6. A flight path planning device for unmanned aerial vehicles (UAVs), characterized in that, include: The data acquisition unit is used to acquire a set of project scene resources and construct point cloud data from the set of project scene resources to obtain a set of point cloud target image data. A time control unit is used to configure the time axis track of the point cloud target image data set to obtain a time axis control data set. A safety constraint unit is used to set safety constraints using the time axis control data set to obtain a flight constraint data set. The path editing unit is used to perform path editing processing on the flight constraint data set to generate an initial flight path data set in a custom timeline; The path settlement unit is used to perform flight path settlement processing based on the initial flight path data set to obtain the target flight path data set. The data debugging unit is used to perform debugging processing on the target flight path data set in a three-dimensional simulation environment to obtain a debugged flight path data set. The path output unit is used to export and deploy the debug flight path data set to obtain a control script data set. The time control unit is specifically used to load a timeline component and construct an editing interface based on the point cloud target image data set to generate a timeline interface data set; configure the time precision of the timeline interface data set to obtain a configured timeline interface data set; configure the UAV position track based on the configured timeline interface data set to obtain a UAV position track data set; configure a light track using the position track data set to bind light on / off and color parameters to track events to obtain a UAV light track data set; and configure a sound effect track based on the light track data set and parse the audio to load the audio onto the track and extract beat points to obtain a timeline control data set. The path editing unit is specifically used to initialize the flight constraint data set with a custom timeline to group the drones and obtain a drone group data set; construct a storyboard interface for the drone group data set to obtain a storyboard image identifier data set; bind point clouds using the storyboard image identifier data set to obtain a storyboard point cloud image data set; configure the timeline of the storyboard point cloud image data set to obtain storyboard bound flight constraint data; and perform time point visualization adjustments based on the storyboard bound flight constraint data to generate an initial flight path data set.

7. A computer device, characterized in that, The device includes a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the UAV flight path planning method as described in any one of claims 1 to 5.

8. 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 UAV flight path planning method as described in any one of claims 1 to 5.

Images