An unmanned aerial vehicle monitoring flight control method, system, device and storage medium

Abstract

The embodiment of the application discloses a kind of unmanned plane monitoring flight control methods, applied to unmanned plane system, it is characterized in that, comprising: the space area information of site is acquired and monitoring task information;Based on the space area information, a plurality of three-dimensional monitoring pipeline models for unmanned plane flight are constructed, and the plurality of three-dimensional monitoring pipeline models correspond to different monitoring target areas or different monitoring task scenes respectively;Real-time situation information of competition is acquired, and the current monitoring target is determined according to the real-time situation information of competition;According to the current monitoring target, the target three-dimensional monitoring pipeline model is determined from the plurality of three-dimensional monitoring pipeline models;Real-time flight state information of unmanned plane is acquired, and based on the real-time flight state information, unmanned plane is controlled to switch to the target three-dimensional monitoring pipeline model and fly in, to realize the monitoring of the current monitoring target.

Description

Technical Field

[0001] This invention relates to the field of intelligent control and event monitoring technology for unmanned aerial vehicles (UAVs), and in particular to a UAV monitoring flight control method, system, device, and storage medium. Background Technology

[0002] With the continuous development of drone platforms, airborne gimbals, visual recognition, and intelligent flight control technologies, drones have been gradually applied to scenarios such as live sports broadcasts, training analysis, security patrols, and multi-angle image acquisition. Especially in football matches, due to the large playing area, frequent personnel movement, rapid transitions between offense and defense, and sudden occurrences of critical events, traditional fixed cameras, while providing stable footage, still fall short in terms of viewing angle flexibility, area tracking capabilities, and rapid response to critical events. Drones, with their advantages of mobility, diverse viewing angles, and flexible deployment, can effectively supplement the shortcomings of ground-based camera systems.

[0003] However, existing drone monitoring and control methods for football matches typically have the following problems: Firstly, flight path planning is mostly based on a single preset route or a simple target tracking strategy, which makes it difficult to take into account multiple monitoring needs such as midfield panorama, half-field advancement, key events in the penalty area, and sideline tracking. Secondly, when the situation in the game changes rapidly, the existing solutions often cannot switch flight strategies in time according to the real-time game status, which can easily cause delays in perspective switching, loss of targets, or poor continuity of images. Third, in actual stadium environments, there are restrictions such as no-fly zones, densely populated areas of players, and safety height constraints, making it difficult for existing drone control methods to strike a balance between safety constraints and monitoring quality. Fourth, some solutions only focus on changes in the drone's position, lacking coordinated control of flight altitude, flight speed, heading, and gimbal orientation, resulting in insufficient stability, completeness, and event representation in the captured footage.

[0004] Therefore, there is an urgent need to propose a drone monitoring flight control scheme suitable for football match scenarios, enabling drones to intelligently switch and fly stably in multiple pre-built flight monitoring spaces according to the real-time situation of the match, thereby improving the continuity, relevance and security of match monitoring. Summary of the Invention

[0005] The purpose of this invention is to provide a method, system, device and storage medium for monitoring and controlling unmanned aerial vehicles (UAVs), so as to at least solve the problems of the prior art, such as the single monitoring strategy of UAVs, untimely response to changes in the competition situation, inaccurate monitoring of key areas, and difficulty in balancing flight safety and field of view continuity during switching.

[0006] To address the aforementioned technical problems, the embodiments of the present invention provide the following technical solutions: A method for monitoring and controlling flight of an unmanned aerial vehicle (UAV) system, applied to an UAV system, includes: Acquire spatial area information of the competition venue and monitoring task information; Based on the spatial region information, multiple three-dimensional monitoring pipeline models for UAV flight are constructed, with each three-dimensional monitoring pipeline model corresponding to a different monitoring target area or a different monitoring task scenario. Acquire real-time situation information of the match and determine the current monitoring target based on the real-time situation information of the match; Based on the current monitoring target, determine the target 3D monitoring pipeline model from the plurality of 3D monitoring pipeline models; The system acquires real-time flight status information of the UAV and controls the UAV to switch to flight within the target 3D monitoring pipeline model based on the real-time flight status information, so as to achieve monitoring of the current monitoring target.

[0007] In some optional embodiments, the method further includes: the spatial area information includes at least one of the following: field boundary information, midfield area information, penalty area information, sideline area information, area behind the goal, stands area information, and no-fly zone information.

[0008] In some optional embodiments, the plurality of three-dimensional monitoring pipeline models include at least two of the following: a panoramic monitoring pipeline model corresponding to full midfield monitoring; a half-field tracking pipeline model corresponding to attacking half-field tracking monitoring; a penalty area monitoring pipeline model corresponding to key events in the penalty area monitoring; a sideline tracking pipeline model corresponding to local tracking of the sideline area; and a goal observation pipeline model corresponding to key observation of the goal area.

[0009] In some optional embodiments, the real-time situation information of the game includes at least one of the following: real-time location information of the monitored target, direction of movement information of the monitored target, player distribution information, offensive and defensive direction information, and status information of key events.

[0010] In some optional embodiments, determining the current monitoring target based on the real-time situation information of the competition includes: determining the target monitoring area based on the real-time location information of the monitoring target; or determining the target event monitoring area based on the status information of key events; or determining the target attack and defense monitoring area based on the attack and defense direction information.

[0011] In some optional embodiments, determining the target three-dimensional monitoring pipeline model based on the current monitoring target includes: matching the current monitoring target with a preset pipeline model mapping relationship; and determining the target three-dimensional monitoring pipeline model corresponding to the current monitoring target based on the matching result.

[0012] In some optional embodiments, the preset pipeline model mapping relationship includes: when the current monitoring target is a midfield global monitoring target, determining the panoramic monitoring pipeline model as the target three-dimensional monitoring pipeline model; when the current monitoring target is an attacking half-field monitoring target, determining the half-field tracking pipeline model as the target three-dimensional monitoring pipeline model; when the current monitoring target is a penalty area key event monitoring target, determining the penalty area monitoring pipeline model as the target three-dimensional monitoring pipeline model; and when the current monitoring target is a sideline tracking monitoring target, determining the sideline tracking pipeline model as the target three-dimensional monitoring pipeline model.

[0013] In some optional implementations, controlling the UAV to switch to flight within the target 3D monitoring pipeline model includes: determining whether the current 3D monitoring pipeline model of the UAV is consistent with the target 3D monitoring pipeline model; if they are inconsistent, generating a switching flight trajectory from the current 3D monitoring pipeline model to the target 3D monitoring pipeline model; and correcting the UAV's flight control commands based on the switching flight trajectory.

[0014] In some optional implementations, the method further includes: continuously acquiring updated real-time situational information of the competition while the UAV flies along the target three-dimensional monitoring pipeline model; when the updated real-time situational information of the competition indicates a change in the current monitoring target, re-determining the target three-dimensional monitoring pipeline model and executing pipeline switching control again.

[0015] Furthermore, the present invention also provides a drone monitoring flight control system, comprising: The information acquisition module is used to acquire spatial area information of the competition venue, monitoring task information, real-time competition situation information, and real-time flight status information of drones. The pipeline construction module is used to construct multiple three-dimensional monitoring pipeline models based on the spatial region information. The target determination module is used to determine the current monitoring target based on the real-time situation information of the competition, and to determine the target's three-dimensional monitoring pipeline model based on the current monitoring target; The flight control module is used to control the UAV to switch to flight within the target 3D monitoring pipeline model based on the real-time flight status information, so as to achieve monitoring of the current monitoring target.

[0016] The present invention also provides an electronic 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 steps of the above method.

[0017] The present invention also provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the steps of the above-described method.

[0018] The present invention has at least the following beneficial effects: By constructing multiple 3D monitoring pipeline models based on the spatial area information of the competition venue, the UAV flight space is transformed from traditional discrete waypoint control to structured flight channel control tailored to the needs of event monitoring, improving the scenario adaptability of flight strategies. By acquiring real-time situational information of the competition and dynamically determining the current monitoring target, UAVs can automatically switch to more suitable monitoring pipelines according to the progress of the competition, thereby improving the response speed to key scenarios such as offensive and defensive transitions, events in restricted areas, and sideline advances. Attached Figure Description

[0019] One or more embodiments are illustrated by way of example with reference numerals in the accompanying drawings. These illustrations do not constitute a limitation on the embodiments. Elements with the same reference numerals in the drawings are denoted as similar elements. Unless otherwise stated, the figures in the drawings are not to be limited by scale.

[0020] Figure 1 This is a schematic diagram of a drone monitoring and flight control method provided in Embodiment 1 of the present invention; Figure 2 This is a flowchart illustrating the transition connection segment provided in Embodiment 1 of the present invention; Figure 3 This is a flowchart illustrating the process of determining the current monitoring target based on the real-time situation information of the competition, provided in Embodiment 1 of the present invention. Figure 4 This is a schematic diagram of a competition scene provided in Embodiment 1 of the present invention; Figure 5 This is a schematic diagram of the structure of a drone monitoring flight control device provided in Embodiment 2 of the present invention; Figure 6 This is a schematic diagram of an electronic device structure provided in Embodiment 3 of the present invention. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, 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 embodiments of the present invention, not all embodiments. 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.

[0022] The technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0023] This embodiment provides a method for monitoring and controlling the flight of an unmanned aerial vehicle (UAV), applied to an UAV system. The UAV system may include one or more of the following: a UAV platform, an onboard camera device, a gimbal control device, a flight controller, a ground control terminal, and a data processing platform.

[0024] like Figure 1 As shown, the method includes the following steps: Step S101: Obtain spatial area information of the competition venue and monitoring task information; Spatial area information can be obtained through manual pre-setting, importing electronic maps of the competition venue, visual recognition mapping, lidar scanning mapping, RTK mapping, or multi-source fusion.

[0025] The spatial area information may include, but is not limited to: field boundary information, midfield area information, penalty area information, sideline area information, area behind the goal information, stands area information, and no-fly zone information.

[0026] Furthermore, the spatial area information can be expanded to include information such as the substitute bench area, highly sensitive areas for referee activities, fixed areas for broadcast equipment, areas with obstacles on site, and emergency avoidance buffer areas, in order to improve the system's adaptability to real-world event scenarios.

[0027] The monitoring task information is used to indicate the monitoring responsibilities of the drones in this event, and may include at least one of the following: panoramic monitoring task, partial follow-up shooting task, key event monitoring task, and area patrol task.

[0028] For example, during the pre-match warm-up phase, monitoring tasks can be biased towards area patrol and panoramic monitoring tasks; during the match, monitoring tasks can be biased towards local follow-up and key event monitoring tasks; and during halftime or post-match, monitoring tasks can be switched to stand order patrol or goal area inspection tasks.

[0029] By acquiring the spatial area information and monitoring task information, basic environmental constraints and task objective constraints can be provided for subsequent UAV flight path planning and monitoring perspective planning.

[0030] S102: Construct multiple three-dimensional monitoring pipeline models for UAV flight based on the spatial area information.

[0031] The three-dimensional monitoring pipeline model can be understood as a three-dimensional flight channel pre-planned above the competition venue, with spatial boundary constraints and monitoring intent constraints.

[0032] The 3D monitoring pipeline model can be expressed using polygonal cylinders, curved swept volumes, mesh volumes, spline envelopes, or tubular regions formed by splicing multiple constrained sections.

[0033] Each 3D monitoring pipeline model may include at least one or more of the following information: horizontal coverage range, vertical height range, allowed flight direction, recommended flight speed range, gimbal pitch / yaw angle range, corresponding monitoring object type, priority or activation condition, and the multiple 3D monitoring pipeline models may correspond to different monitoring target areas or different monitoring task scenarios.

[0034] Specifically, the multiple three-dimensional monitoring pipeline models may include at least two of the following: a panoramic monitoring pipeline model, mainly covering a higher, safer area above the midfield or the entire field, used to provide overall game situation monitoring and tactical panoramic views. Its flight altitude can be relatively high, and its flight speed can be relatively low to obtain a stable, wide-angle view. A half-field tracking pipeline model is mainly set up above or to the side of a specific offensive or defensive half of the field, used to quickly track the direction of ball movement during offensive and defensive transitions, suitable for capturing the process of advancing from midfield into the penalty area. The half-field tracking pipeline is positioned at a lower altitude than the panoramic monitoring pipeline.

[0035] The penalty area monitoring pipeline model, primarily set up around the penalty area and its perimeter, can be used to monitor key events such as shots, saves, aerial duels, corner kicks, and penalty kicks. This pipeline model requires high stability of the viewing angle and strong target focusing capabilities. The sideline tracking pipeline model extends along the sideline of the field and is suitable for local tracking in scenarios such as wing breakthroughs, sideline kicks, and fast counter-attacks. The goal observation pipeline model is set up in the safe monitoring airspace near the goal area for focused observation of the goal area, suitable for judging shot trajectories, goalmouth melees, and goalkeeper activity.

[0036] In some alternative embodiments, "transitional connection segments" or "switching buffer pipes" can be pre-built between different 3D monitoring pipeline models to enable the UAV to switch more smoothly between multiple main monitoring pipeline models.

[0037] Specifically, the transition connection segment can be set between two adjacent or task-switching 3D monitoring pipeline models to form a spatial connection channel. The transition connection segment can be constructed using a curve connection structure, a spline surface connection structure, a gradually varying cross-section channel structure, or a spatial channel structure generated based on trajectory interpolation, so that when the UAV enters the second monitoring pipeline model from the first monitoring pipeline model, its flight trajectory can meet the continuity constraint and dynamic feasibility constraint.

[0038] In some embodiments, the transition connection segment can be generated in the following manner: S1021: Determine the outlet and inlet sections of the two target monitoring pipeline models; S1022: Based on the spatial relationship between the outlet section and the inlet section, a smooth connection path is generated using a three-dimensional spline curve, Bezier curve, B-spline curve, or minimum curvature path algorithm. S1023: Using the connection path as the central axis, a transition pipe region with certain width and height constraints is generated through spatial expansion or envelope, thereby forming the switching buffer pipe.

[0039] Specifically, the end section of the first monitored pipeline model can be obtained as the outlet section, and the starting section of the second monitored pipeline model can be obtained as the inlet section. The section can be determined by the center point of the section, the normal vector of the section, and the boundary contour of the section, wherein the boundary contour can be represented by a polygonal contour or a circular contour. Furthermore, the spatial coordinates, height information, and pipeline cross-sectional dimensions of the outlet and inlet sections can also be recorded.

[0040] Based on the spatial positional relationship between the exit section and the entrance section, spatial geometric parameters between them are calculated. These parameters may include: the spatial distance between the center points of the two sections, the height difference between the two sections, the directional angle between the two sections, and the horizontal offset between the two sections. After obtaining these spatial geometric parameters, a smooth flight path connecting the two sections is generated using a path generation algorithm.

[0041] In some embodiments, to ensure the smoothness of the UAV flight trajectory, curvature continuity constraints and acceleration continuity constraints can be set during the path generation process so that the generated path meets the dynamic constraints of the UAV flight control system.

[0042] After obtaining the smooth connection path, the connection path is used as the central axis of the transition pipe, and the transition pipe region is generated by spatial expansion or envelope.

[0043] Specifically, the center point of the exit section can be used as the starting point of the path, and the center point of the entrance section as the ending point. The starting and ending tangent vectors of the path are set according to the current flight direction of the UAV and the entry direction of the target monitoring pipeline. Based on this, the connecting path can be generated using at least one of the following path generation methods: 3D spline curve interpolation algorithm, Bezier curve generation algorithm, B-spline curve fitting algorithm, and minimum curvature path generation algorithm based on curvature constraints.

[0044] In some embodiments, to ensure the smoothness of the UAV flight trajectory, curvature continuity constraints and acceleration continuity constraints can be set during the path generation process so that the generated path meets the dynamic constraints of the UAV flight control system.

[0045] After obtaining the smooth connection path, the connection path is used as the central axis of the transition pipe, and the transition pipe region is generated by spatial expansion or envelope.

[0046] Specifically, multiple path sampling points can be generated along the central axis at preset sampling intervals, and a cross-sectional constraint region can be established at each sampling point. The cross-sectional constraint region can be represented by a circular cross-section, an elliptical cross-section, or a polygonal cross-section, and its cross-sectional dimensions can be set according to the UAV's safe flight distance, attitude adjustment space, and navigation error tolerance.

[0047] In some embodiments, the cross-sectional dimensions can be dynamically changed according to the path location, for example: A larger cross-sectional dimension is set near the outlet of the original monitoring pipeline to provide greater space for attitude adjustment; Maintain stable cross-sectional dimensions in the middle region of the path; The cross-sectional dimensions are gradually reduced near the inlet of the target monitoring pipeline to achieve a smooth connection.

[0048] Subsequently, by connecting the sampling sections, a continuous three-dimensional envelope structure can be formed, thereby generating a transition pipe region with certain width and height constraints.

[0049] Furthermore, to ensure the flight safety of the UAV in the transition pipe, flight constraint parameters can be set for the transition pipe area, including: maximum permissible flight speed, maximum permissible turning angular velocity, maximum permissible pitch angle change rate, maximum permissible altitude change rate, and gimbal attitude adjustment range.

[0050] By using the above parameter constraints, it can be ensured that the UAV can adjust its flight direction, speed, and monitoring perspective in the transition channel, thereby achieving a smooth transition between monitoring tasks.

[0051] The transition pipe generated in the above manner can form a continuous, smooth spatial connection structure between two monitoring pipe models that satisfies flight control constraints, thereby enabling the UAV to achieve stable path switching when performing different monitoring tasks and improving the continuity and stability of the monitoring image.

[0052] Step S103: Obtain real-time situation information of the match, and determine the current monitoring target based on the real-time situation information of the match; The real-time situation information of the game may include at least one of the following: real-time location information of the monitored target, direction of movement information of the monitored target, player distribution information, offensive and defensive directions, and status information of key events.

[0053] The real-time situation information of the match can be obtained from: airborne visual recognition results, player positioning equipment or football positioning chip data, edge computing node analysis results, manual marking information, and match event stream data.

[0054] Among them, the status information of key events may include status information such as free kicks, corner kicks, penalty kicks, throw-ins, counter-attacks, scrambles in front of the goal, offside disputes, goalkeeper saves, and celebration gatherings.

[0055] Determining the current monitoring target based on the real-time situation information of the match can include one or more of the following methods: determining the target monitoring area based on the real-time position information of the football, determining the target event monitoring area based on the status information of key events, and determining the target offensive and defensive monitoring area based on the offensive and defensive direction information. For example: when the football is near the midfield and the formations of both sides are relatively balanced, the current monitoring target can be determined as the overall midfield monitoring target; when the football is rapidly advancing towards a certain half of the field, the current monitoring target can be determined as the attacking half monitoring target; when the football enters the penalty area or a key event such as a corner kick or penalty kick is identified, the current monitoring target can be determined as the penalty area key event monitoring target; when the football is close to the sideline and is continuously advancing, the current monitoring target can be determined as the sideline tracking monitoring target.

[0056] In some implementations, a target scoring mechanism may also be introduced when determining the current monitoring target.

[0057] For example, importance score, urgency score, persistence score, and image value score are calculated for multiple candidate monitoring targets, and the target with the highest combined score is selected as the current monitoring target, thereby improving decision stability.

[0058] In some alternative implementations, acquiring real-time match situation information and determining the current monitoring target based on the real-time match situation information further includes the following steps.

[0059] S1031, Acquire image information of the field and identify whether it belongs to the midfield area, penalty area, passing area, or near the goal.

[0060] Specifically, it can acquire image information of a football field and identify the midfield area, penalty area, crossing area, or area near the goal; it can also be image information of a basketball court or a badminton court.

[0061] S1032, Obtain the location information of the monitored targets in the stadium.

[0062] Optionally, the monitoring target can be a ball game in the relevant competition, such as a football game, in which the monitoring target is football; or a basketball game, in which the monitoring target is basketball. Taking a football game as an example, the monitoring target can be football. Optionally, the monitoring target can be player distribution information. By statistically analyzing player aggregation density, densely populated areas of players can be identified as monitoring targets. Based on the player aggregation density and movement direction, it can be determined whether the drone chooses to fly in the global pipeline module or the half-court pipeline model.

[0063] S1033, Based on the location information of the monitored target and combined with the situation prediction model, predict the short-term future competition trend and determine the three-dimensional monitoring pipeline for the drone's flight.

[0064] Specifically, when the monitoring target is a soccer ball, based on the soccer ball's movement direction information, the vector of the soccer ball's position change over consecutive moments can be used to determine whether it is moving towards the left half, the right half, or shifting laterally within a certain area. This determines whether the drone should fly within the global pipeline module or the half-field pipeline model. When the monitoring target is a densely populated area of ​​players, the three-dimensional monitoring pipeline for the drone's flight is determined based on the players' movement directions.

[0065] In some implementations, short-term future match trends can also be predicted based on situation prediction models.

[0066] For example, when a football is detected approaching the edge of the penalty area at high speed and accompanied by an attacking player making a quick run, a critical event in the penalty area can be predicted in advance, and the drone can be switched to the penalty area monitoring pipeline model in advance, thereby improving the timeliness of capturing critical events.

[0067] In some alternative embodiments, when the current 3D monitoring pipeline model of the UAV is inconsistent with the target 3D monitoring pipeline model, the control system generates a switching flight trajectory.

[0068] In some alternative implementations, the flight path switching is divided into two stages: The first stage is the exit from the current pipeline stage, which allows the drone to smoothly exit the current pipeline model; The second stage is the target pipeline access stage, which enables the UAV to enter the target pipeline in an attitude and speed that meet the constraints of the target pipeline.

[0069] A transition buffer zone can be set between the two stages to reduce sharp turns and sudden attitude changes.

[0070] In some alternative embodiments, the parameters of the UAV are adjusted in a coordinated manner according to the type of the 3D pipeline model.

[0071] In this embodiment, flight control is not limited to controlling the position change of the UAV, but can also simultaneously control at least one of the following parameters: flight altitude, flight speed, flight heading, gimbal pitch angle, gimbal yaw angle, lens zoom parameters, image stabilization mode parameters, recommended heading range parameters, recommended gimbal pitch range parameters, pipeline priority parameters, and pipeline activation condition parameters.

[0072] Specifically, in panoramic monitoring tasks, the drone can be controlled to maintain a high flight altitude, low translation speed, and a small gimbal pitch angle to obtain a wide-range panoramic view; in sideline tracking tasks, the drone's flight speed along the sideline direction can be increased to better match the speed of the football's movement, while the gimbal yaw angle can be adjusted to continuously track the ball-carrying area; in penalty area key event monitoring tasks, the flight speed can be appropriately reduced and combined with lens zoom to highlight shots, saves, or multiple players vying for possession.

[0073] By constructing transitional connection segments or switching buffer channels between different monitoring pipeline models, and generating connection paths through a smooth path generation algorithm, the UAV can maintain the continuity of its flight trajectory and the smoothness of its attitude changes when switching between different monitoring pipelines, thereby avoiding sudden changes in flight path or drastic changes in attitude and improving the continuity and stability of the monitoring footage.

[0074] S104: Based on the current monitoring target, determine the target three-dimensional monitoring pipeline model from the plurality of three-dimensional monitoring pipeline models.

[0075] In this step, the current monitoring target can be matched with the preset pipeline model mapping relationship, and the target three-dimensional monitoring pipeline model corresponding to the current monitoring target can be determined based on the matching result.

[0076] For example, the preset pipeline model mapping relationship may include: when the current monitoring target is a midfield global monitoring target, determining the panoramic monitoring pipeline model as the target three-dimensional monitoring pipeline model; when the current monitoring target is an attacking half-field monitoring target, determining the half-field tracking pipeline model as the target three-dimensional monitoring pipeline model; when the current monitoring target is a restricted area key event monitoring target, determining the restricted area monitoring pipeline model as the target three-dimensional monitoring pipeline model; and when the current monitoring target is a sideline tracking monitoring target, determining the sideline tracking pipeline model as the target three-dimensional monitoring pipeline model.

[0077] Furthermore, when determining the target three-dimensional monitoring pipeline model, it can also be jointly determined in conjunction with the monitoring task information.

[0078] For example, when all current monitoring targets are pointing to restricted areas, if the current monitoring task is panoramic monitoring, then the high-altitude restricted area monitoring pipeline with wider coverage will be selected first; if the current monitoring task is close-up monitoring of key events, then the low-speed fine monitoring pipeline that is closer to the target area and has a more focused perspective will be selected first.

[0079] In an optional implementation, the target 3D monitoring pipeline model is not limited to a single model, but can also be determined as a set of candidate monitoring pipelines. Specifically, the system can select several candidate pipeline models that can be used to perform the current monitoring task from multiple 3D monitoring pipeline models based on the current monitoring task requirements, the current location of the UAV, and the spatial coverage relationship of each monitoring pipeline, thereby forming the set of candidate monitoring pipelines.

[0080] Furthermore, after determining the candidate monitoring pipeline set, each candidate pipeline in the set can be comprehensively evaluated, and the candidate pipelines can be ranked according to the evaluation results. The ranking criteria may include one or more of the following evaluation indicators: The pipeline priority index allows different monitoring pipeline models to be pre-set with different priorities based on the needs of event monitoring. For example, the penalty area monitoring pipeline has a higher priority than the panoramic monitoring pipeline during the shooting or corner kick phase, while the panoramic monitoring pipeline can have a relatively higher priority during the overall game situation monitoring phase.

[0081] The estimated image quality can be determined based on the expected flight position of the UAV in the candidate pipeline and the gimbal's viewing angle parameters, thus assessing the image quality of the target area. The estimated image quality can be calculated based on the following factors: The spatial distance between the drone and the target area, whether the gimbal's pitch and yaw angles are within the optimal shooting range, the coverage ratio of the target area in the image, and any possible obstructions or visual interference factors.

[0082] In some embodiments, a shooting quality assessment model can be established to calculate the shooting quality score of each candidate pipeline by weighting the above factors.

[0083] The pipeline switching cost index can also be calculated based on factors such as the transition path length between the UAV's current monitoring pipeline and the candidate monitoring pipeline, flight time, and attitude adjustment requirements. The switching cost can include: path length cost, energy consumption cost, attitude adjustment complexity cost, and flight safety risk cost.

[0084] In some embodiments, the above-mentioned multiple indicators can be fused in a weighted manner to obtain a comprehensive score for each candidate pipeline, for example: Overall score = α × Pipeline priority score + β × Image quality score - γ × Switching cost score; Wherein, α, β, and γ are preset weighting coefficients.

[0085] After obtaining the comprehensive score of each candidate pipeline, the candidate monitoring pipeline set can be sorted according to the score results, and the monitoring pipeline with the highest score can be selected as the target three-dimensional monitoring pipeline model.

[0086] Subsequently, the UAV can generate a flight control path based on the selected target monitoring pipeline model and the corresponding transition connection section, and execute the corresponding flight control commands to achieve high-quality monitoring of the target monitoring area.

[0087] By introducing a candidate monitoring pipeline set and its evaluation and ranking mechanism, drones can automatically select the optimal monitoring path from multiple monitoring pipelines when facing different competition events or monitoring task requirements, thereby improving the flexibility, shooting quality and overall intelligence level of the event monitoring system.

[0088] By constructing a set of candidate monitoring pipelines and comprehensively evaluating and ranking them based on various evaluation indicators such as pipeline priority, estimated image quality, and pipeline switching cost, the UAV can automatically select the optimal monitoring pipeline to execute flight control, thereby improving the intelligent decision-making capability of the monitoring system in complex competition scenarios.

[0089] S105: Obtain the real-time flight status information of the UAV, and control the UAV to switch to flight within the target three-dimensional monitoring pipeline model based on the real-time flight status information.

[0090] The real-time flight status information of the UAV may include at least one of the following: current location, flight altitude, flight speed, flight heading, attitude angle, remaining battery level, remaining flight time, communication link quality, gimbal angle, and onboard sensor status.

[0091] The control of the drone to switch to flight within the target 3D monitoring pipeline model may include: First, determine whether the current 3D monitoring pipeline model of the drone is consistent with the target 3D monitoring pipeline model; If they are consistent, maintain the current flight within the pipeline and make minor adjustments to the flight control commands based on the local target position; If there is a discrepancy, a switching flight trajectory is generated from the current 3D monitoring pipeline model to the target 3D monitoring pipeline model, and the flight control commands of the UAV are corrected according to the switching flight trajectory.

[0092] The switching flight path must meet at least one of the following conditions: avoid no-fly zones, avoid airspace above areas with high concentrations of players, and maintain continuous field-of-view coverage of the current monitored target. Continuous field-of-view coverage can be understood as follows: during the switching from the current channel to the target channel, the drone or its gimbal should, as far as possible, keep the current monitored target within the effective area of ​​the frame to reduce camera jumps and target loss.

[0093] Furthermore, when controlling the UAV to switch to flight within the target 3D monitoring pipeline model, at least one of the UAV's flight altitude, flight speed, flight heading, and gimbal orientation is adjusted in a coordinated manner.

[0094] For example, when switching from panoramic monitoring to monitoring of critical events in restricted areas, the flight altitude can be appropriately reduced, the flight speed decreased, and the gimbal pitch angle adjusted to enhance the ability to capture target details. When switching from sideline tracking to midfield panoramic monitoring, the flight altitude can be appropriately increased, the gimbal's wide-angle coverage range increased, and the local tracking weight reduced to restore the overall situational awareness.

[0095] In an optional embodiment, each three-dimensional monitoring pipeline model can be defined by the following parameters: three-dimensional spatial boundary parameters, target area relative position parameters, minimum safe height parameters, maximum allowable height parameters, optimal shooting distance parameters, and pipeline priority parameters.

[0096] For example, the panoramic monitoring pipeline model can be deployed in a relatively high area above the center of the field, with its width covering the center circle and adjacent areas, suitable for obtaining an overall tactical view; the half-field tracking pipeline model can be set up symmetrically on the left and right sides according to the direction of attack and defense, corresponding to the left and right halves respectively; the penalty area monitoring pipeline model can be set up around the two penalty areas respectively, and can form an arc-shaped observation pipeline in front of the penalty area to adapt to the monitoring of shooting and defensive confrontation; the sideline tracking pipeline model can be deployed parallel to the two sidelines, and allows for local trajectory correction as the position of the football moves; the goal observation pipeline model can be located in a safe area behind or diagonally above the goal to form a more favorable observation angle in front of the goal.

[0097] In some alternative implementations, each 3D monitoring pipeline model may also carry image quality prediction parameters to evaluate the target sharpness, occlusion probability, view integrity, and broadcast value when shooting from the pipeline in the current environment, so as to select a better option from multiple candidate pipelines.

[0098] Compared with existing technologies, the UAV monitoring method and system proposed in this application have the following beneficial effects: They improve the structuring of UAV monitoring path planning by constructing multiple three-dimensional monitoring pipeline models based on the spatial area information of the competition venue, transforming the UAV's flight path from traditional free path planning into a structured monitoring flight channel. This enables stable monitoring of key areas while meeting flight safety constraints, improving the controllability and standardization of UAV flight path planning. Furthermore, by constructing different types of three-dimensional monitoring pipeline models for different competition areas, such as panoramic monitoring pipelines, half-field tracking pipelines, penalty area monitoring pipelines, sideline tracking pipelines, and goal observation pipelines, UAVs can select the most suitable monitoring position and perspective in different competition scenarios, thereby significantly improving the shooting quality and monitoring effect of key areas of the event.

[0099] In one embodiment of this application, a drone monitoring flight control device is also provided. This device can be installed in a drone system, such as in a drone flight control system, a drone mission control module, or a drone ground control terminal, to implement the aforementioned drone monitoring flight control method.

[0100] The UAV monitoring flight control device 500 includes: an acquisition module 501, a pipeline construction module 502, a situation acquisition module 503, a pipeline determination module, and a flight control module.

[0101] The acquisition module is used to acquire spatial area information of the competition venue and monitoring task information.

[0102] Specifically, the spatial area information can be obtained through manual pre-setting, electronic map import, visual recognition mapping, LiDAR scanning mapping, RTK mapping, or multi-source fusion mapping. This spatial area information may include information on the field boundaries, midfield area, penalty area, sideline area, area behind the goal, stands area, and no-fly zones. The monitoring task information is used to indicate the monitoring responsibilities of the drone during the event, such as panoramic monitoring, local follow-up filming, key event monitoring, and area patrol tasks.

[0103] The pipeline construction module is used to construct multiple 3D monitoring pipeline models for UAV flight based on the spatial area information. These 3D monitoring pipeline models can be understood as pre-planned 3D flight channels with spatial boundary constraints above the playing field. Each of the multiple 3D monitoring pipeline models corresponds to a different monitoring target area or a different monitoring task scenario, such as a panoramic monitoring pipeline model, a half-field tracking pipeline model, a penalty area monitoring pipeline model, a sideline tracking pipeline model, and a goal observation pipeline model.

[0104] The situation acquisition module is used to acquire real-time situation information of the match and determine the current monitoring target based on the real-time situation information. The real-time situation information can be obtained through match video analysis, player and monitoring target position detection, event data systems, or sensor data. For example, when a football is detected entering the penalty area, the event in the penalty area can be identified as the current monitoring target; when a breakthrough is detected on the wing, the event in the sideline area can be identified as the current monitoring target.

[0105] The pipeline determination module is used to determine a target 3D monitoring pipeline model from among the multiple 3D monitoring pipeline models based on the current monitoring target. Specifically, the pipeline determination module can select the most suitable target monitoring pipeline for performing the current monitoring task from among the multiple 3D monitoring pipeline models based on information such as the area location corresponding to the current monitoring target, monitoring viewpoint requirements, and the current position of the UAV.

[0106] The flight control module acquires real-time flight status information of the UAV and controls the UAV to switch to flight within the target 3D monitoring pipeline model based on this information, thereby enabling monitoring of the current target. The real-time flight status information may include the UAV's current position, flight altitude, flight speed, flight attitude, and remaining battery power. The flight control module can generate flight control commands based on the UAV's current flight status and the positional relationship with the target 3D monitoring pipeline model, thereby controlling the UAV to enter the target 3D monitoring pipeline model and fly along the pipeline.

[0107] In some embodiments, when a UAV needs to switch from the current monitoring pipeline to the target 3D monitoring pipeline model, the system can also generate a smooth transition flight path through a transition connection segment or a switching buffer pipeline, thereby enabling the UAV to switch smoothly between different monitoring pipelines, so as to avoid sudden changes in flight path and improve the stability of the monitoring image.

[0108] The aforementioned device structure enables drones to automatically select appropriate monitoring flight channels and execute flight control based on the real-time situation of the match, thereby achieving high-quality monitoring of key areas or key events in football matches, improving the intelligence level of the event monitoring system and the stability of drone flight control.

[0109] According to embodiments of the present invention, an electronic device is provided, such as... Figure 6The diagram shown is a structural schematic of an electronic device according to Embodiment 3 of the present invention. This electronic device may include a processor 601, a communication interface 602, a memory 603, and a communication bus 604. The processor 601, communication interface 602, and memory 603 communicate with each other via the communication bus 604. The processor 601 can call logical instructions stored in the memory 603. The processor is used to execute computer programs stored in the memory to achieve the following functions: First, acquire spatial area information and monitoring task information for the match venue. The spatial area information may include information on the football field boundaries, midfield area, penalty area, sideline area, area behind the goal, stands, and no-fly zones. The monitoring task information can be used to instruct the drone's monitoring responsibilities during the event, such as panoramic monitoring, local follow-up filming, key event monitoring, or area patrol.

[0110] Then, based on the spatial area information, multiple three-dimensional monitoring pipeline models for UAV flight are constructed. These multiple three-dimensional monitoring pipeline models correspond to different monitoring target areas or different monitoring task scenarios, such as panoramic monitoring pipeline models, half-field tracking pipeline models, restricted area monitoring pipeline models, sideline tracking pipeline models, and goal observation pipeline models.

[0111] Subsequently, real-time match situation information is acquired, and the current monitoring target is determined based on this information. This real-time match situation information can be obtained through video analysis, event data systems, or target detection algorithms, such as detecting the position of the football, player movement trajectories, or key match events.

[0112] Next, based on the current monitoring target, the target three-dimensional monitoring pipeline model is determined from the multiple three-dimensional monitoring pipeline models.

[0113] After determining the target 3D monitoring pipeline model, the real-time flight status information of the UAV is obtained, and flight control commands are generated based on the real-time flight status information to control the UAV to switch to flight within the target 3D monitoring pipeline model, so as to realize the monitoring of the current monitoring target.

[0114] The real-time flight status information may include the drone's spatial location, flight altitude, flight speed, attitude angle, and battery status.

[0115] In some embodiments, when the UAV needs to switch from the current monitoring pipeline to the target 3D monitoring pipeline model, the processor can also generate a transition flight path based on the spatial relationship between the current monitoring pipeline and the target monitoring pipeline, thereby enabling the UAV to smoothly switch between different 3D monitoring pipeline models.

[0116] The communication interface 602 is used to realize data communication between electronic devices and other modules of the UAV or ground control system, such as receiving real-time situation information of the competition, sending flight control commands, and synchronously monitoring task information.

[0117] In some embodiments, the processor may be implemented using a central processing unit (CPU), a digital signal processor (DSP), a graphics processing unit (GPU), a field-programmable gate array (FPGA), or an application-specific integrated circuit (ASIC).

[0118] Through the aforementioned electronic devices, drone monitoring and flight control based on the real-time situation of the competition can be realized, enabling the drone to automatically select a suitable monitoring flight channel and execute flight control, thereby improving the automation level of the competition monitoring system and the stability of drone flight control. Furthermore, the logical instructions in the aforementioned memory 603 can be implemented as software functional units and, when sold or used as independent products, can be stored in several computer-readable storage media. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in Embodiment 1 of the present invention. The aforementioned storage media include: USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, optical disks, and other media capable of storing program code.

[0119] The above-mentioned electronic device can execute any of the target annotation methods described in Embodiment 1, and has the corresponding functional modules and beneficial effects of the method. For technical details not described in detail in this embodiment, please refer to the target annotation method provided in Embodiment 1 of the present invention.

[0120] According to an embodiment of the present invention, a computer-readable storage medium of the type described in Embodiment 3 is provided. The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor performs the steps of the target annotation method described in Embodiment 1.

[0121] The above-mentioned products can execute any of the target annotation methods described in Embodiment 1, and have the corresponding functional modules and beneficial effects of the method. For technical details not described in detail in this embodiment, please refer to the target annotation method provided in Embodiment 1 of this invention.

[0122] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a general-purpose hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the related technology, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., including several instructions to enable a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in various embodiments or some parts of the embodiments.

[0123] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; under the concept of the present invention, the technical features of the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations of different aspects of the present invention as described above, which are not provided in detail for the sake of brevity; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A method for monitoring and controlling the flight of an unmanned aerial vehicle (UAV), applied to an UAV system, characterized in that, Acquire spatial area information of the competition venue and monitoring task information; Based on the spatial region information, multiple three-dimensional monitoring pipeline models for UAV flight are constructed, with each three-dimensional monitoring pipeline model corresponding to a different monitoring target area or a different monitoring task scenario. Acquire real-time situation information of the match and determine the current monitoring target based on the real-time situation information of the match; Based on the current monitoring target, a target 3D monitoring pipeline model is determined from the multiple 3D monitoring pipeline models. Based on the current monitoring task requirements, the current location of the UAV, and the spatial coverage relationship of the multiple 3D monitoring pipeline models, a candidate monitoring pipeline set is determined. After determining the candidate monitoring pipeline set, the candidate monitoring pipelines are comprehensively evaluated to obtain a comprehensive score for the candidate pipelines. The candidate monitoring pipeline set is then sorted according to the score results, and the 3D monitoring pipeline with the highest score is selected as the target 3D monitoring pipeline model. Acquire real-time flight status information of the UAV, and control the UAV to switch to flight within the target 3D monitoring pipeline model based on the real-time flight status information; The multiple 3D monitoring pipeline models include at least two of the following: A panoramic monitoring pipeline model corresponding to the midfield panoramic monitoring; A half-court tracking pipeline model corresponding to offensive half-court tracking and monitoring; A restricted area monitoring pipeline model corresponding to the monitoring of critical events in restricted areas; The pipe model that corresponds to the local tracking of the edge line area; A model of the goal observation channel corresponding to the key observation area of ​​the goal; The control of the UAV to switch to flight within the target 3D monitoring pipeline model includes: Determine whether the current 3D monitoring pipeline model of the UAV is consistent with the target 3D monitoring pipeline model; In case of inconsistency, a switching flight trajectory is generated from the current 3D monitoring pipeline model to the target 3D monitoring pipeline model; The flight control commands for the UAV are corrected based on the changed flight trajectory.

2. The UAV monitoring and flight control method according to claim 1, characterized in that, The spatial area information includes at least one of the following: field boundary information, midfield area information, penalty area information, sideline area information, area behind the goal, stands area information, and no-fly zone information.

3. The UAV monitoring and flight control method according to claim 1, characterized in that, The real-time situation information of the game includes at least one of the following: real-time location information of the monitored target, direction of movement information of the monitored target, player distribution information, offensive and defensive direction information, and status information of key events.

4. The UAV monitoring and flight control method according to claim 1, characterized in that, The step of determining the current monitoring target based on the real-time situation information of the match includes: Determine the target monitoring area based on the real-time location information of the monitored target; And / or determine the target event monitoring area based on key event status information; And / or determine the target attack and defense monitoring area based on attack and defense direction information.

5. The UAV monitoring and flight control method according to claim 1, characterized in that, The step of determining the target 3D monitoring pipeline model based on the current monitoring target includes: Match the current monitoring target with the preset pipeline model mapping relationship; Based on the matching results, determine the target three-dimensional monitoring pipeline model corresponding to the current monitoring target.

6. The UAV monitoring and flight control method according to claim 5, characterized in that, The preset pipeline model mapping relationship includes: When the current monitoring target is the overall monitoring target in the field, the panoramic monitoring pipeline model is determined to be the target three-dimensional monitoring pipeline model; When the current monitoring target is the attack half-court monitoring target, the half-court tracking pipeline model is determined to be the target three-dimensional monitoring pipeline model; When the current monitoring target is a key event monitoring target in a restricted area, the restricted area monitoring pipeline model is determined to be a three-dimensional monitoring pipeline model of the target. When the current monitoring target is a sideline tracking monitoring target, the sideline tracking pipeline model is determined to be the target three-dimensional monitoring pipeline model.

7. The UAV monitoring and flight control method according to claim 1, characterized in that, Also includes: As the drone flies along the target 3D monitoring pipeline model, it continuously acquires updated real-time situational information of the competition. When the updated real-time situation information of the game indicates a change in the current monitoring target, the target's three-dimensional monitoring pipeline model is redefined, and pipeline switching control is executed again.

8. An electronic device, characterized in that, include: processor; Memory; and computer programs stored in the memory and executable on the processor, The processor, when executing the computer program, implements the UAV monitoring and flight control method as described in any one of claims 1-7.

Images