A method and system for real-time fine adjustment of a UAV route with local adjustment support
By dragging waypoints or path segments on a graphical user interface, smooth local transition paths can be generated and detected in real time, solving the problems of inflexible UAV flight path adjustment and computational redundancy. This improves the flexibility and intuitiveness of UAV flight path adjustment and increases operational efficiency.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- NANJING QUFEIPAI TECHNOLOGY CO LTD
- Filing Date
- 2026-03-17
- Publication Date
- 2026-06-05
AI Technical Summary
Existing drone flight path adjustments are inflexible, operations are not intuitive, and calculations are redundant, resulting in low operational efficiency and an inability to quickly respond to temporary obstacles or changes in mission details.
By dragging waypoints or path segments directly on the graphical user interface, the system generates smooth local transition paths in real time, performs real-time conflict detection and dynamic feasibility verification, generates local update commands and sends them to the UAV to achieve local flight path adjustments.
It enables flexible adjustment of drone flight paths, improves operational intuitiveness, reduces computational redundancy, and enhances operational efficiency and continuity.
Smart Images
Figure CN122151893A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of unmanned aerial vehicle (UAV) flight control technology, and more specifically, to a method and system for real-time fine-tuning of UAV flight paths that supports local adjustments. Background Technology
[0002] Automated flight path (AWACS) for drones is widely used in surveying, inspection, agricultural plant protection, and aerial filming. The standard operating procedure is as follows: the user pre-plans the complete flight path (including waypoints, maneuvers, flight parameters, etc.) in the Ground Control Station (GCS) software, then uploads it to the drone and executes it.
[0003] However, existing technologies have the following drawbacks: Inflexible adjustments: When the mission is executed, adjustments to the flight path are required due to temporary obstacles (such as newly appearing cranes or trees), weather changes, or changes in mission details (such as the need to shoot from a specific angle). Existing solutions typically require pausing the current mission, returning to the flight path planning interface, modifying the global flight path, recalculating, and uploading it. This process is cumbersome and time-consuming, severely impacting operational efficiency and continuity.
[0004] The operation is not intuitive: modifying flight routes often requires users to input precise latitude and longitude coordinates or complex parameters, which is extremely unfriendly for rapid on-site response.
[0005] Computational redundancy: Even if only one waypoint in the route is modified, the system often needs to recalculate and verify the entire route, resulting in unnecessary computational overhead.
[0006] No effective solutions have yet been proposed to address the problems in the relevant technologies. Summary of the Invention
[0007] The main purpose of this application is to provide a method and system for real-time fine-tuning of UAV flight paths that supports local adjustments, in order to solve the problems of inflexible adjustments, unintuitive operation, and redundant calculations.
[0008] To achieve the above objectives, according to one aspect of this application, a method for real-time fine-tuning of UAV flight paths that supports local adjustments is provided.
[0009] The real-time fine-tuning method for UAV flight paths supporting local adjustments according to this application includes the following steps: Display the drone's current flight path on the UI and set unexecuted future segments to an interactive state; In response to a user's drag operation on a waypoint or route segment in the future flight segment, determine the affected local adjustment area; Using the starting and ending points of the local adjustment area as fixed boundaries, and combining the new path points after dragging, a smooth local transition path is generated in real time and previewed on the UI. The waypoint data corresponding to the local transition path is encapsulated into a local update command and sent to the UAV; The UAV receives and loads the local update instruction, and automatically switches to the local transition path when flying to the starting point of the local adjustment area.
[0010] Furthermore, the UI preview also includes: Real-time conflict detection is performed on the local transition path; If the conflict detection passes, the waypoint data corresponding to the local transition path is encapsulated into a local update command and sent to the UAV.
[0011] Furthermore, the local adjustment area is a continuous segment centered on the towed waypoint, with at least one waypoint before and after it.
[0012] Furthermore, using the starting and ending points of the local adjustment area as fixed boundaries, and combining the new path points after dragging, a smooth local transition path is generated in real time, including: A piecewise cubic Hermit spline interpolation method is adopted, with the starting and ending points of the local adjustment area as fixed boundaries, and combined with the new path points after dragging, to generate a smooth local transition path in real time. The local transition path is then optimized by a lightweight trajectory optimization method based on quadratic programming.
[0013] Furthermore, real-time conflict detection of the local transition path includes: Detect and acquire the real-time status of the drone; By calling upon a high-precision digital elevation model and obstacle database, and combining the real-time status of the UAV, rapid geometrical collision detection is performed on local transition paths. If the rapid collision detection in geometric space passes, then the dynamic feasibility of the local transition path is verified. If the fast collision detection in geometric space fails, a conflict resolution suggestion will be provided.
[0014] Furthermore, after performing a dynamic feasibility check on the local transition path, the following steps are also included: If the dynamic feasibility verification passes, a local update instruction is generated; If the dynamic feasibility verification fails, user adjustment suggestions will be provided.
[0015] Furthermore, when the UAV receives and loads the local update instruction, and automatically switches to the local transition path upon reaching the starting point of the local adjustment area, the process also includes: The drone uses forward-looking local perception to detect obstacles in real time and obtain real-time information. The system compares real-time obstacle information with models in the obstacle database and high-precision digital elevation models, and triggers dynamic obstacle avoidance strategies based on the comparison results. The dynamic obstacle avoidance strategies include: pausing and issuing an alarm or dynamic fine-tuning.
[0016] To achieve the above objectives, according to another aspect of this application, a real-time fine-tuning system for unmanned aerial vehicle (UAV) flight paths that supports local adjustments is provided.
[0017] The real-time fine-tuning system for drone flight paths supporting partial adjustments according to this application includes: The settings module is used to display the drone's current flight path on the UI and to set unexecuted future segments to an interactive state; The response module is used to respond to the user's drag operation on a waypoint or path segment in the future flight segment and determine the affected local adjustment area; The generation module is used to generate a smooth local transition path in real time, based on the starting and ending points of the local adjustment area as fixed boundaries and combined with the new path points after dragging, and to preview it on the UI. The encapsulation module is used to encapsulate the waypoint data corresponding to the local transition path into a local update command and send it to the UAV; The switching module is used by the UAV to receive and load the local update command, and automatically switch to the local transition path when flying to the starting point of the local adjustment area.
[0018] In this embodiment, a real-time fine-tuning method for UAV flight paths that supports local adjustments is adopted. By dragging and dropping local segments of the flight path that is being executed or about to be executed directly through a graphical user interface, the system only re-plans the path for the affected part of the flight path and dynamically sends the planned flight path to the UAV. This eliminates the need to interrupt the task or re-plan the global flight path, achieving flexible adjustment, intuitive operation, and reduced computational redundancy. Attached Figure Description
[0019] The accompanying drawings, which form part of this application, are used to provide a further understanding of the application and to make other features, objects, and advantages of the application more apparent. The illustrative embodiments and descriptions of this application are used to explain the application and do not constitute an undue limitation of the application. In the drawings: Figure 1 This is a flowchart illustrating the method for real-time fine-tuning of drone flight paths according to this application; Figure 2 This is a schematic diagram of the structure of the real-time fine-tuning system for unmanned aerial vehicle (UAV) flight paths according to this application. Detailed Implementation
[0020] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, and not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative effort should fall within the scope of protection of the present application.
[0021] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate for the embodiments of this application described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.
[0022] In this application, the terms "upper," "lower," "left," "right," "front," "rear," "top," "bottom," "inner," "outer," "middle," "vertical," "horizontal," "lateral," and "longitudinal" indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. These terms are primarily for the purpose of better describing the invention and its embodiments, and are not intended to limit the indicated device, element, or component to having a specific orientation, or to be constructed and operated in a specific orientation.
[0023] Furthermore, in addition to indicating direction or positional relationship, some of the aforementioned terms may also have other meanings. For example, the term "above" may also be used in certain situations to indicate a dependency or connection. Those skilled in the art can understand the specific meaning of these terms in this invention based on the specific circumstances.
[0024] Furthermore, the terms "installation," "setup," "equipped with," "connection," "linking," and "socketing" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral structure; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium, or an internal connection between two devices, components, or parts. Those skilled in the art can understand the specific meaning of these terms in this invention based on the specific circumstances.
[0025] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.
[0026] According to embodiments of the present invention, a method is provided, such as Figure 1 As shown, the method includes the following steps S101 to S104: Step S101: Display the current flight path of the drone on the UI and set unexecuted future flight segments to an interactive state; Specifically, on the client's UI map interface, the entire flight path of the current task is displayed as a highlighted line.
[0027] Completed flight segments are displayed in gray, currently flying segments are displayed in a special color (such as blue), and future flight segments that have not yet flown are displayed in an editable state (such as solid green lines or dashed lines).
[0028] During operation, users can directly use the mouse or their finger to long-press and drag any waypoint on the future flight segment. While dragging, the waypoint and its adjacent affected paths will be previewed in real time as a variable curve. Users can also directly drag the path segment between two waypoints, and the system will automatically insert a temporary waypoint at the mouse position and perform the drag operation.
[0029] Step S102: Respond to the user's drag operation on a waypoint or path segment in the future flight segment and determine the affected local adjustment area; Specifically, the system continuously monitors for drag events in real time, meaning it constantly listens for drag gestures from the user on editable future flight segments. When the user begins to drag a waypoint (P), the system automatically locks the flight segment centered on P and consisting of N waypoints before and after it (e.g., N=2) as a local adjustment zone. The flight path outside the adjustment zone is considered fixed.
[0030] Step S103: Using the starting and ending points of the local adjustment area as fixed boundaries, and combining the new path points after dragging, generate a smooth local transition path in real time, and preview it on the UI. In this embodiment, preferably, generating a smooth local transition path in real time by using the starting and ending points of the local adjustment area as fixed boundaries and combining the new path points after dragging includes: using piecewise cubic Hermit spline interpolation, using the starting and ending points of the local adjustment area as fixed boundaries and combining the new path points after dragging, generating a smooth local transition path in real time, and optimizing the local transition path by a lightweight trajectory optimization method of quadratic programming.
[0031] The specific process of piecewise cubic Hermitian spline interpolation is as follows: The preferred approach is segmented construction: the starting point P_start of the local adjustment area, the drag point P_drag generated after user operation, and the ending point P_end of the local adjustment area are sequentially used as the key nodes of the path. A tangent vector (guide vector) is estimated for each node. For P_start and P_end, their tangent vectors are the heading vectors V_start and V_end of the original route. For P_drag, its tangent vector can be automatically estimated using the Catmull-Rom spline rule: the normalized result of T_drag = (P_drag - P_start) + (P_end - P_drag), or determined by solving a small system of linear equations.
[0032] Then, a spline is generated: between every two adjacent nodes (e.g., from P_start to P_drag, from P_drag to P_end), a cubic Hermitian spline segment is uniquely determined using the positions of the two endpoints and the tangent vector. The parametric equations of this segment are simple in form, have extremely low computational cost, and guarantee C1 continuity (position and tangent continuity) at the connection points.
[0033] Finally, curvature smoothing: If higher-order C2 continuity (curvature continuity) is required, cubic B-splines (B-Splines) or non-uniform rational B-splines (NURBS) can be used.
[0034] The specific method uses the derivative information of P_start, P_drag, P_end, and boundary points as control points or constraints, and generates globally C2 continuous splines by solving a tridiagonal linear equation system. This equation system is very small (usually involving only 3-5 points), and the solution speed still meets real-time requirements.
[0035] The specific process of the lightweight trajectory optimization method of quadratic programming is as follows: First, path parameterization: The path from P_start to P_end is discretized vertically into N equally spaced "path points" {s_i}, where s_1 = P_start and s_N = P_end. N is a small value (e.g., 5-10) to ensure real-time performance. The lateral offsets {l_i} of these path points are used as optimization variables. The initial offset corresponds to the original route (l_i = 0).
[0036] Then, we construct an optimization problem: the objective function is to minimize the "unsmoothness" of the path and the deviation from the target point.
[0037] min Σ( (l_{i-1} - 2*l_i + l_{i+1})^2 ) + w * (l_k - l_drag)^2 The first term is an approximation of the square of curvature (to smooth the path), the second term requires the k-th path point to be close to the drag target point, and w is the weight.
[0038] Boundary constraints: l_1 = 0, l_N = 0 (no lateral offset at the start and end points).
[0039] Dynamic constraint: |l_{i-1} - 2*l_i + l_{i+1}| ≤ κ_max (upper limit of curvature constraint).
[0040] Finally, a real-time solution is obtained: The optimization problem described above is a standard convex quadratic programming (QP) problem with a very small size. It can be solved in 1-5 milliseconds using an efficient solver (such as an OSQP-based solver library), yielding a set of optimal lateral offsets {l_i*}. Combined with the original longitudinal path, a smooth optimized path can be generated.
[0041] In the UI interaction thread, Option 1 (Hermite spline) is used first for the real-time preview of the first stage, with almost no delay and providing a silky smooth drag-and-drop feel.
[0042] Before the user releases the mouse for final confirmation, Option 2 (QP optimization) can be triggered in parallel to achieve finer path adjustments that take into account dynamic constraints, and the optimized path can be used as the final result.
[0043] In this embodiment, preferably, after previewing on the UI, the method further includes: Real-time conflict detection is performed on the local transition path; If the conflict detection passes, the waypoint data corresponding to the local transition path is encapsulated into a local update command and sent to the UAV.
[0044] Since the local transition path is replanned, real-time conflict detection needs to be performed on the local transition path before it is sent. This allows for a quick and accurate determination of whether the new path is safe the moment the user drags and drops to generate a preview path, and provides immediate feedback.
[0045] Specifically, real-time conflict detection of the local transition path includes: Detect and acquire the real-time status of the drone; By calling upon a high-precision digital elevation model and obstacle database, and combining the real-time status of the UAV, rapid geometrical collision detection is performed on local transition paths. If the rapid collision detection in geometric space passes, then the dynamic feasibility of the local transition path is verified. If the fast collision detection in geometric space fails, a conflict resolution suggestion will be provided.
[0046] After performing a dynamic feasibility check on the local transition path, the following steps are also included: If the dynamic feasibility verification passes, a local update instruction is generated; If the dynamic feasibility verification fails, user adjustment suggestions will be provided.
[0047] The specific process for fast collision detection in geometric space is as follows: The newly generated smooth local transition path (curve) is discretized into a series of dense detection points with a fixed step size (e.g., 0.5-1 meter).
[0048] The system maintains a 3D spatial index (such as an octree, KD tree, or 3D grid) based on preloaded data. To improve the speed of local queries, this index can be organized by geographic blocks.
[0049] For each detection point, a tiny "detection cylinder" or "detection cube" is constructed with that point as the center, based on the drone's safety radius (body size + margin) and preset safety height (such as above the minimum safety height of terrain / buildings).
[0050] Use spatial indexing to quickly query whether there are any static obstacles (terrain elevation points, building voxels, etc.) within the range of the "detection body".
[0051] Simultaneously, the latitude and longitude coordinates of the detection point are quickly compared with the predefined no-fly / restricted-fly zone polygons for inclusion determination. If a path point falls into a no-fly zone or its altitude is below the restricted-fly zone requirement, it is immediately determined to be a conflict.
[0052] There are no conflicts at any of the detection points, and the preview path is displayed in a "safe" color (such as green).
[0053] If any detection point triggers a conflict, the point and the affected path segment will immediately be highlighted on the UI in a "warning" color (such as red or flashing orange), and a concise text prompt (such as "Conflict with known building") may pop up.
[0054] This algorithm ensures millisecond-level continuous operation during user dragging, providing instant visual feedback.
[0055] The specific process for verifying the kinetic feasibility is as follows: Calculate the curvature change of the local transition path to ensure that its minimum turning radius is greater than the minimum executable turning radius of the drone at the current flight speed. If the path is too sharp, it is considered a dynamic conflict, and the user is prompted to reduce speed or adjust the path.
[0056] Calculate the maximum slope angle of the path segment and ensure it is less than the drone's maximum permissible climb / descent angle. This prevents the drone from stalling or overloading due to excessive vertical dragging by the user.
[0057] When the UAV receives and loads the local update instruction, and automatically switches to the local transition path upon reaching the starting point of the local adjustment area, the flight path further includes: The drone uses forward-looking local perception to detect obstacles in real time and obtain real-time information. The system compares real-time obstacle information with models in the obstacle database and high-precision digital elevation models, and triggers dynamic obstacle avoidance strategies based on the comparison results. The dynamic obstacle avoidance strategies include: pausing and issuing an alarm or dynamic fine-tuning.
[0058] As the drone flies toward the starting point of the adjustment zone, its onboard sensors (such as forward-looking binocular vision and lidar) continuously scan the local new path area ahead.
[0059] The real-time perceived point cloud data is compared with the static model used in the first layer of detection. If a new major obstacle is detected (whose location and size are sufficient to threaten flight safety) and the obstacle is within the safety envelope of the new path, a dynamic obstacle avoidance strategy is immediately triggered.
[0060] The dynamic obstacle avoidance strategy is as follows: The system immediately hovers and sends a high-level alarm to the ground station, alerting the user to the presence of an unknown obstacle. This achieves the purpose of pausing and issuing an alarm.
[0061] By combining more sophisticated online path planning algorithms (such as Fast Exploratory Random Tree (RRT)*), and using a safe point in front as the target, a very short detour path is generated to guide the UAV to avoid dynamic obstacles before returning to the predetermined local path. This achieves the goal of online fine-tuning. Furthermore, when a conflict is detected, the system not only issues an alert but also provides intelligent solutions to enhance the user experience.
[0062] Along the normal direction of the conflict point (upward or horizontally), automatically calculate the nearest safe point that satisfies all constraints (such as minimum gap and spatial rules), and preview a new automatically adjusted path for the user to choose from. This achieves automatic offset.
[0063] Parameter relaxation suggestion: prompt the user to "try raising the waypoint by X meters" or "move Z meters in the Y direction to avoid the obstacle".
[0064] Step S104: Encapsulate the waypoint data corresponding to the local transition path into a local update command and send it to the UAV; Step S105: The UAV receives and loads the local update instruction, and automatically switches to the local transition path when flying to the starting point of the local adjustment area.
[0065] The generated local transition path is encapsulated as an instruction and sent to the drone. The drone then flies according to the instruction.
[0066] like Figure 2 As shown, this application also relates to a real-time fine-tuning system for UAV flight paths supporting local adjustments, comprising: a setting module 10, used to display the current flight path of the UAV on the UI and set unexecuted future segments to an interactive state; a response module 20, used to respond to a user's drag operation on a waypoint or path segment in the future segment and determine the affected local adjustment area; a generation module 30, used to generate a smooth local transition path in real time, using the start and end points of the local adjustment area as fixed boundaries and combining the new path points after dragging, and preview it on the UI; an encapsulation module 40, used to encapsulate the waypoint data corresponding to the local transition path into a local update instruction and send it to the UAV; and a switching module 50, used for the UAV to receive and load the local update instruction, and automatically switch to the local transition path when flying to the start point of the local adjustment area. This achieves the same technical effect as the real-time fine-tuning method for UAV flight paths supporting local adjustments.
[0067] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.
Claims
1. A method for real-time fine-tuning of UAV flight paths supporting local adjustments, characterized in that, Includes the following steps: Display the drone's current flight path on the UI and set unexecuted future segments to an interactive state; In response to a user's drag operation on a waypoint or route segment in the future flight segment, determine the affected local adjustment area; Using the starting and ending points of the local adjustment area as fixed boundaries, and combining the new path points after dragging, a smooth local transition path is generated in real time and previewed on the UI. The waypoint data corresponding to the local transition path is encapsulated into a local update command and sent to the UAV; The UAV receives and loads the local update instruction, and automatically switches to the local transition path when flying to the starting point of the local adjustment area.
2. The method for real-time fine-tuning of UAV flight paths according to claim 1, characterized in that, The preview on the UI also includes: Real-time conflict detection is performed on the local transition path; If the conflict detection passes, the waypoint data corresponding to the local transition path is encapsulated into a local update command and sent to the UAV.
3. The method for real-time fine-tuning of UAV flight paths according to claim 1, characterized in that, The local adjustment area is a continuous segment centered on the towed waypoint, with at least one waypoint before and after it.
4. The method for real-time fine-tuning of UAV flight paths according to claim 1, characterized in that, Using the starting and ending points of the local adjustment area as fixed boundaries, and combining the new path points after dragging, a smooth local transition path is generated in real time, including: A piecewise cubic Hermit spline interpolation method is adopted, with the starting and ending points of the local adjustment area as fixed boundaries, and combined with the new path points after dragging, to generate a smooth local transition path in real time. The local transition path is then optimized by a lightweight trajectory optimization method based on quadratic programming.
5. The method for real-time fine-tuning of UAV flight paths according to claim 2, characterized in that, Real-time collision detection of the local transition path includes: Detect and acquire the real-time status of the drone; By calling upon a high-precision digital elevation model and obstacle database, and combining the real-time status of the UAV, rapid geometrical collision detection is performed on local transition paths. If the rapid collision detection in geometric space passes, then the dynamic feasibility of the local transition path is verified. If the fast collision detection in geometric space fails, a conflict resolution suggestion will be provided.
6. The method for real-time fine-tuning of UAV flight paths according to claim 5, characterized in that, After performing a dynamic feasibility check on the local transition path, the following steps are also included: If the dynamic feasibility verification passes, a local update instruction is generated; If the dynamic feasibility verification fails, user adjustment suggestions will be provided.
7. The method for real-time fine-tuning of UAV flight paths according to claim 6, characterized in that, When the UAV receives and loads the local update instruction, and automatically switches to the local transition path upon reaching the starting point of the local adjustment area, the flight path further includes: The drone uses forward-looking local perception to detect obstacles in real time and obtain real-time information. The system compares real-time obstacle information with models in the obstacle database and high-precision digital elevation models, and triggers dynamic obstacle avoidance strategies based on the comparison results. The dynamic obstacle avoidance strategies include: pausing and issuing an alarm or dynamic fine-tuning.
8. A real-time fine-tuning system for UAV flight paths that supports local adjustments, characterized in that, include: The settings module is used to display the drone's current flight path on the UI and to set unexecuted future segments to an interactive state; The response module is used to respond to the user's drag operation on a waypoint or path segment in the future flight segment and determine the affected local adjustment area; The generation module is used to generate a smooth local transition path in real time, based on the starting and ending points of the local adjustment area as fixed boundaries and combined with the new path points after dragging, and to preview it on the UI. The encapsulation module is used to encapsulate the waypoint data corresponding to the local transition path into a local update command and send it to the UAV; The switching module is used for the UAV to receive and load the local update instruction, and automatically switch to the local transition path when flying to the starting point of the local adjustment area.