A collaborative control system and method for intelligent remote sensing targets based on unmanned aerial vehicles (UAVs)

Abstract

This application provides an intelligent remote sensing target cooperative control system and method based on unmanned aerial vehicles (UAVs), relating to the field of target cooperative control technology. By segmenting and registering pose interaction attributes with all thermal tracking markers, adjacency correction information is obtained when the UAV performs heading registration on the remote sensing mapping path. Then, based on the positioning constraint strategy when the UAV locates the target along the flight path and all texture scan markers, the cooperative planning level of the UAV during active mapping tilt angle planning is determined. The cooperative planning level is then reset in a linked manner to obtain the point-to-point reset trajectory when the UAV responds to target position adjustments in the mapping area. Quantitative mapping control is performed on the mapping targets within the mapping area based on the aerial survey viewpoint coordinates and the point-to-point reset trajectory. This application can dynamically register the target cooperative control process in UAV remote sensing mapping tasks within an intelligent remote sensing target cluster environment, thereby improving the UAV's mapping response capability to ground targets.

Description

Technical Field

[0001] This application relates to the field of target collaborative control technology, and more specifically, to an intelligent remote sensing target collaborative control system and method based on unmanned aerial vehicles (UAVs). Background Technology

[0002] Target cooperative control refers to the coordination of the response behavior and control modes of various targets in multi-target mapping or remote sensing missions, taking into account their spatial distribution and response characteristics. This allows for coordinated operation of position, attitude, and response signals throughout the overall mission execution, thereby improving the data acquisition accuracy and target recognition efficiency of the remote sensing system. In UAV-based intelligent remote sensing scenarios, target cooperative control can dynamically adjust the reflectivity, visibility angle, or signal response mode of targets based on the UAV's flight path, dynamic perspective, and imaging rhythm. This enables multiple targets to simultaneously possess high recognizability and geometric solvability in remote sensing imaging, making it particularly suitable for mapping and tracking missions in complex terrain or large areas, achieving unified control and efficient feedback of multi-target information.

[0003] However, existing UAV-based intelligent remote sensing target cooperative control methods lack a dynamic recognition mechanism for the remote sensing target's state. This prevents the UAV from achieving real-time pose interaction with the ground target during remote sensing mapping, leading to heading registration errors and tilt angle response lag during aerial survey path planning. Consequently, the accuracy of target positioning and the integrity of mapping data are reduced. Therefore, how to dynamically register the target cooperative control process in UAV remote sensing mapping missions within an intelligent remote sensing target swarm environment to improve the UAV's mapping response capability to ground targets is a challenge facing the industry. Summary of the Invention

[0004] This application provides a collaborative control system and method for intelligent remote sensing targets based on unmanned aerial vehicles (UAVs). It can dynamically register the collaborative control process of targets in UAV remote sensing mapping tasks in an intelligent remote sensing target cluster environment, so as to improve the UAV's mapping response capability to ground targets.

[0005] In a first aspect, this application provides a collaborative control method for intelligent remote sensing targets based on unmanned aerial vehicles (UAVs), the collaborative control method comprising the following steps:

[0006] Multiple sets of self-identifying intelligent remote sensing targets are deployed in the surveying area, and thermal tracking and texture scanning marks are set on each set of targets for remote sensing measurements by UAVs.

[0007] Determine the pose interaction attributes of the UAV when it flies along the flight path in the monitoring area, and perform segmented registration of the pose interaction attributes with all thermal tracking markers to obtain the adjacency correction information when the UAV performs heading registration on the remote sensing mapping path. Then, extract the aerial survey viewpoint coordinates of the UAV when performing aerial survey calibration on the target from the adjacency correction information.

[0008] Based on the positioning constraint strategy when the UAV locates the target along the flight path and all texture scan marks, the collaborative planning level of the UAV is determined when actively planning the mapping tilt angle. The collaborative planning level is then reset in a coordinated manner to obtain the point reset trajectory of the UAV when it responds to the adjustment of the target position in the mapping area.

[0009] Quantitative mapping control is performed on the mapping targets within the mapping area based on the aerial survey viewpoint coordinates and the point reset trajectory.

[0010] In this embodiment, determining the pose interaction attributes of the UAV while it flies along the flight path within the monitoring area specifically includes:

[0011] To obtain the pose evolution trend of the UAV as it flies along the flight path;

[0012] Based on the pose evolution trend, determine the interactive calibration features of the UAV when performing mapping and calibration within the monitoring area;

[0013] The pose interaction attributes of the UAV when flying along the flight path within the monitoring area are determined based on the interaction calibration features.

[0014] In this embodiment, the pose interaction attribute refers to the matching information between the attitude change pattern of the UAV during flight and the target mapping requirements.

[0015] In this embodiment, the adjacency correction information refers to the set of smooth transition parameters for attitude adjustment between adjacent track segments.

[0016] In this embodiment, extracting the aerial survey viewpoint coordinates of the UAV during aerial survey calibration of the target from the adjacency correction information specifically includes:

[0017] Based on the adjacency correction information, a calibration compensation rule is constructed for the UAV to perform aerial survey calibration on the target;

[0018] The calibration compensation rules are mapped to the UAV's onboard camera coordinate system to generate viewpoint attitude features;

[0019] The coordinates of the aerial survey viewpoint when the UAV performs aerial survey calibration on the target are determined based on the viewpoint attitude characteristics.

[0020] In this embodiment, determining the collaborative planning level of the UAV during active planning of the mapping tilt angle, based on the positioning constraint strategy when the UAV performs target localization along the flight path and all texture scan identifiers, specifically includes:

[0021] Obtain the positioning constraint strategy when the UAV locates a target along its flight path;

[0022] Determine the planning efficiency boundary of the UAV in active planning of the mapping tilt angle based on all texture scan identifiers;

[0023] The collaborative planning level of the UAV in active planning of surveying tilt angle is determined by the positioning constraint strategy and the planning efficiency boundary.

[0024] In this embodiment, the active mapping tilt angle planning refers to a planning method in which the UAV dynamically adjusts the camera's shooting tilt angle based on terrain and target features to improve the quality of remote sensing data.

[0025] In this embodiment, the quantitative mapping control of the mapping target within the mapping area based on the aerial survey viewpoint coordinates and the sub-point reset trajectory specifically includes:

[0026] The attitude contribution of the survey target is determined based on the coordinates of the aerial survey viewpoint.

[0027] The dynamic tracking index of the survey target within the survey area is generated by the point reset trajectory;

[0028] The target mapping control strategy within the mapping area is verified by the attitude contribution and the dynamic tracking index, and adaptive orientation matching is performed.

[0029] In this embodiment, the positioning constraint strategy refers to the rules that restrict the spatial attitude, flight trajectory, and target recognition angle of the UAV during the execution of the mapping task.

[0030] Secondly, this application provides a UAV-based intelligent remote sensing target cooperative control system for executing a UAV-based intelligent remote sensing target cooperative control method, the cooperative control system comprising:

[0031] The label setting module is used to deploy multiple sets of self-identifying intelligent remote sensing targets in the surveying area, and to set thermal tracking labels and texture scanning labels on each set of targets when the UAV performs remote sensing measurements;

[0032] The segmented registration module is used to determine the pose interaction attributes of the UAV when it flies along the flight path in the monitoring area, and to perform segmented registration of the pose interaction attributes with all thermal tracking markers to obtain the adjacency correction information when the UAV performs heading registration on the remote sensing mapping path. Then, the aerial survey viewpoint coordinates of the UAV when performing aerial survey calibration on the target are extracted from the adjacency correction information.

[0033] The linkage reset module is used to determine the collaborative planning level of the UAV during active planning of the mapping tilt angle based on the positioning constraint strategy and all texture scan marks when the UAV locates the target along the flight path. The collaborative planning level is then linked to reset to obtain the point reset trajectory of the UAV when it responds to the adjustment of the target position in the mapping area.

[0034] The quantitative control module is used to perform quantitative mapping control on the mapping targets within the mapping area based on the aerial survey viewpoint coordinates and the point reset trajectory.

[0035] The technical solutions provided by the embodiments disclosed in this application have the following beneficial effects:

[0036] Multiple sets of self-identifying intelligent remote sensing targets are deployed within the surveying area. Thermal tracking markers and texture scanning markers are set on each target for UAV remote sensing measurements. The pose interaction attributes of the UAV as it flies along its flight path within the monitoring area are determined. These pose interaction attributes are then segmented and registered with all thermal tracking markers to obtain adjacency correction information for UAV heading registration on the remote sensing surveying path. The aerial survey viewpoint coordinates for UAV aerial calibration of the targets are extracted from this adjacency correction information. Based on the positioning constraint strategy for UAV target positioning along its flight path and all texture scanning markers, the collaborative planning level for active planning of the surveying tilt angle is determined. This collaborative planning level is then reset to obtain the point-by-point reset trajectory for UAV to adjust the target position within the surveying area. Quantitative surveying control of the surveying targets within the surveying area is then performed based on the aerial survey viewpoint coordinates and the point-by-point reset trajectory.

[0037] Therefore, this application demonstrates a significant improvement in the dynamic recognition and registration capabilities of UAVs for ground-based intelligent targets, addressing the inaccuracy of target coordination responses in existing UAV remote sensing mapping processes. Specifically, by deploying multiple sets of self-identifying intelligent remote sensing targets within the mapping area and assigning thermal tracking and texture scanning markers to each set, dual-channel recognition of the target targets in both thermal infrared and spatial image data is achieved, enhancing the stable tracking performance and recognition accuracy of UAVs in complex terrain. Furthermore, by determining the pose interaction attributes of the UAV while flying along its flight path within the monitoring area and performing segmented registration with all thermal tracking markers, dynamic corrections can be made for errors caused by inaccurate target coordination during flight. To address attitude deviations caused by environmental disturbances, adjacency correction information is generated for heading registration, thereby improving the geometric continuity and registration robustness of remote sensing mapping paths. By determining the collaborative planning hierarchy based on positioning constraint strategies and texture scanning identifiers along the UAV flight path and implementing linked reset, dynamic planning of multi-point flight tilt angles and observation paths can be performed for target positions, improving flight safety and planning efficiency during tilt angle adjustment. By combining aerial survey viewpoint coordinates and point reset trajectories for quantitative mapping control of survey targets, high-precision point reconstruction and dynamic error compensation within the survey area can be achieved, effectively improving the spatial accuracy of remote sensing modeling and the consistency of multi-view fusion.

[0038] In summary, the technical solution adopted in this application can dynamically register the target collaborative control process in UAV remote sensing mapping tasks in an intelligent remote sensing target cluster environment, thereby improving the UAV's mapping response capability to ground targets. Attached Figure Description

[0039] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only for this embodiment of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0040] Figure 1 This is an exemplary flowchart of a collaborative control method for intelligent remote sensing targets based on unmanned aerial vehicles (UAVs) provided in this application.

[0041] Figure 2 This is a flowchart illustrating the process for determining adjacency correction information provided in this application;

[0042] Figure 3 This is a flowchart illustrating the determination of the point reset trajectory provided in this application;

[0043] Figure 4 This is a module structure diagram of an intelligent remote sensing target collaborative control system based on unmanned aerial vehicles (UAVs) provided in this application. Detailed Implementation

[0044] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0045] This application provides a collaborative control system and method for intelligent remote sensing targets based on unmanned aerial vehicles (UAVs). The core of this system involves deploying multiple sets of self-identifying intelligent remote sensing targets within a surveying area. Each target set is equipped with thermal tracking markers and texture scanning markers for UAV remote sensing measurements. The system determines the pose interaction attributes of the UAV while flying along its flight path within the monitoring area. These pose interaction attributes are then segmented and registered with all thermal tracking markers to obtain adjacency correction information for UAV heading registration on the remote sensing surveying path. Furthermore, the system extracts the aerial survey viewpoint coordinates of the UAV when performing aerial surveying calibration on the targets from the adjacency correction information. Based on the positioning constraint strategy for UAV target positioning along its flight path and all texture scanning markers, the system determines the collaborative planning level for active planning of the surveying tilt angle. The collaborative planning level is then reset in a coordinated manner to obtain the point-by-point reset trajectory of the UAV in response to target position adjustments within the surveying area. Finally, quantitative surveying control is performed on the surveying targets within the surveying area based on the aerial survey viewpoint coordinates and the point-by-point reset trajectory.

[0046] Example 1: To better understand the above technical solution, the following will provide a detailed description of the technical solution in conjunction with the accompanying drawings and specific implementation methods. (Refer to...) Figure 1 As shown in the figure, this is an exemplary flowchart of a collaborative control method for intelligent remote sensing targets based on unmanned aerial vehicles (UAVs) according to this embodiment of the present application. The collaborative control method includes the following steps:

[0047] In step S1, multiple sets of self-identifying intelligent remote sensing targets are deployed in the survey area, and thermal tracking marks and texture scanning marks are set on each set of targets for remote sensing measurements by UAVs.

[0048] In practice, deploying multiple sets of self-identifying intelligent remote sensing targets within the surveying area can be achieved in the following way: First, select a target structure with thermal emission capabilities and high-contrast texture patterns. The thermal sensing part uses a constant-temperature heating element or infrared emitting unit, and the texture pattern uses a checkerboard, QR code, or high-frequency line pattern, forming an intelligent remote sensing target with both visual and thermal infrared recognition capabilities. Then, based on the terrain features of the surveying area, the required surveying resolution, and the flight path of the UAV, spatial planning of the deployment points is performed, typically using a regular grid or triangular mesh distribution to ensure uniform coverage. Finally, using a total station or a high-precision global navigation satellite system, three-dimensional coordinate calibration is completed at each deployment point, and the target location information is entered into the surveying task database, thus completing the deployment of the self-identifying intelligent remote sensing targets.

[0049] It should be noted that, in this application, the self-identifying intelligent remote sensing target is a ground reference object used for remote sensing mapping, which has the ability to be automatically identified and actively provide location information.

[0050] In addition, in practical implementation, the thermal tracking and texture scanning markers for UAV remote sensing measurements on each set of targets can be achieved in the following way: A constant-temperature infrared heating unit or thermal radiation patch is embedded in the central area of ​​the target. The temperature can be set within the range of 40 to 60 degrees Celsius based on expert experience or experimentation, ensuring a clear heat source image is formed under the infrared thermal imager, thus constituting the thermal tracking marker. Secondly, a high-contrast, asymmetrical texture pattern, such as a black and white checkerboard, radial rings, or coded graphics, is laid on the target surface, enabling the visible light or multispectral camera on the UAV to quickly locate and match the target area in the image. To improve recognition efficiency, these marker patterns need to be optimized using image feature extraction algorithms to ensure they possess attributes such as clear corners and stable edges, thus constituting the texture scanning marker.

[0051] It should be noted that, in this application, thermal tracking identifier refers to a constant-temperature heat source area identified by an infrared sensor; texture scanning identifier refers to an image attribute label extracted from a ground texture image that characterizes the intensity and recognizability of regional texture features.

[0052] In step S2, the pose interaction attributes of the UAV when flying along the track in the monitoring area are determined, and the pose interaction attributes are segmented and registered with all thermal tracking markers to obtain the adjacency correction information when the UAV performs heading registration on the remote sensing mapping path. Then, the aerial survey viewpoint coordinates when the UAV performs aerial survey calibration on the target are extracted from the adjacency correction information.

[0053] In this embodiment, determining the pose interaction attributes of the UAV while it flies along a flight path within the monitoring area can be achieved through the following steps:

[0054] To obtain the pose evolution trend of the UAV as it flies along the flight path;

[0055] Based on the pose evolution trend, determine the interactive calibration features of the UAV when performing mapping and calibration within the monitoring area;

[0056] The pose interaction attributes of the UAV when flying along the flight path within the monitoring area are determined based on the interaction calibration features.

[0057] In practice, firstly, during flight, the UAV utilizes an inertial navigation system, a global navigation satellite system, and a flight control system to jointly collect position and attitude data in real time. Position data includes three-dimensional coordinates, and attitude data includes pitch, yaw, and roll angles. The sampling frequency can be set to at least 50 Hz based on expert experience. Then, the continuously sampled data is input sequentially into the flight control data analysis module, where a continuous attitude trajectory curve is formed using trajectory fitting methods (such as spline interpolation or B-spline curve fitting). Simultaneously, the rate of attitude change is differentiated to obtain derivative information such as angular velocity and angular acceleration, which are used to characterize the dynamic trend of the flight state. The processed position and attitude data serve as the attitude evolution trend of the UAV as it flies along the flight path. Then, during the image acquisition period, the corresponding pose data is extracted, and the recognition accuracy score of the remote sensing target in the image recognition module is called simultaneously. Combined with indicators such as the pixel offset of the target center point, observation angle, and image clarity in the aerial survey image, a correspondence matrix between attitude and image acquisition quality is established. Furthermore, attitude points with high image acquisition success rate or optimal image clarity are extracted from the UAV, and the attitude parameters corresponding to these extracted points are used as interactive calibration features for the UAV during mapping and calibration within the monitoring area. Finally, the interactive calibration features are compared point-by-point with the real-time attitude data along the UAV's current flight path, calculating the differences in attitude angle, heading, and target observation angle. Based on the feature matching degree of the comparison results, the flight process is divided into high-matching, medium-matching, and low-matching zones. High-matching areas are directly recorded as flight segments that have met the pose interaction conditions; for medium and low-matching areas, the flight control parameters (such as increasing the pitch angle or changing the yaw angle) are adjusted to simulate feature consistency adjustment, thereby correcting the pose state on the flight path, and all feature matching degrees are used as interactive calibration features when the UAV performs mapping and calibration in the monitoring area.

[0058] It should be noted that, in this application, the pose evolution trend refers to the process of continuous change in the spatial position and attitude angle of the UAV during flight; the interactive calibration feature refers to the spatial registration relationship between the UAV's own pose and the target position when the UAV observes the remote sensing target during flight; and the pose interaction attribute refers to the matching information between the UAV's attitude change pattern and the target mapping requirements during flight.

[0059] Preferably, in this embodiment, the pose interaction attribute is segmented and registered with all thermal tracking identifiers to obtain adjacency correction information when the UAV performs heading registration on the remote sensing mapping path, with reference to... Figure 2 As shown in the figure, this is a schematic flowchart of determining adjacency correction information in some embodiments of this application. In this embodiment, determining adjacency correction information can be achieved by the following steps:

[0060] In step S21, the track segment intervals on the remote sensing mapping path are divided according to the pose interaction attributes;

[0061] In step S22, the tracking correction index of the UAV when performing heading registration on the remote sensing mapping path is extracted from the track segment interval;

[0062] In step S23, the pose matching deviation of the UAV when performing target recognition is determined based on all thermal tracking markers;

[0063] In step S24, the adjacency correction information for the UAV when performing heading registration on the remote sensing mapping path is determined based on the tracking correction index and the pose matching deviation.

[0064] In practice, the process begins by reading the three-dimensional spatial coordinates, pitch angle, yaw angle, roll angle, and calibration score related to image acquisition performance from the pose interaction attributes. Then, a segmentation threshold is set. If the attitude angle change exceeds a certain set angle (e.g., 5 degrees) or the image clarity decreases within three consecutive time windows, it can be considered a new segmentation starting point. A sliding window method is then used to scan the entire path data, automatically identifying breakpoints according to the set threshold rules. These breakpoints divide the path into several track segment intervals, maintaining relative consistency in flight attitude, image acquisition performance, and target observation conditions within each segment. Next, statistical analysis is performed on the pose interaction attributes within each segment, extracting indicators such as attitude angle change gradient, image center offset, and image clarity decline trend. For example, the image center offset can be calculated by identifying the centroid of the target's thermal region in each frame and performing Euclidean distance calculations with the image center point; the attitude angle change gradient is derived from the difference between the pitch and yaw angles in adjacent time periods. The obtained indicators are organized into time series data in chronological order, and their changing trends are calculated. The changing trend can be calculated using the slope method. Points with a rate of change exceeding the warning threshold are identified as key correction locations, and their corresponding time, spatial coordinates, and angle deviations are recorded to form the tracking correction indicators for UAV heading registration on the remote sensing mapping path. Then, the UAV, equipped with an infrared thermal imaging device, continuously captures thermal images of ground targets during flight. The thermal tracking markers detected in each frame are spatially back-projected to calculate the position of the heat source center point in the image coordinate system. This position is then compared with the preset ideal imaging viewpoint (i.e., the optimal observation attitude defined in the interactive calibration features) to analyze the yaw angle error, pitch angle shift, and image shift direction under the current flight attitude. A three-dimensional attitude error calculation model is used to quantify the deviation between the actual UAV pose and the optimal recognition pose in each recognition action, and the timestamp and spatial position are recorded to form the pose matching deviation for UAV target recognition. Finally, for each pair of adjacent track segments, the tracking correction indicators and pose matching deviations at the end and start points are compared. If the pose matching deviation exceeds a preset threshold, the segment is considered to have an abrupt change. A smoothing interpolation method (such as cubic spline interpolation or spherical linear interpolation) can be used to calculate a transition segment between the preceding and following segments, generating multiple intermediate pose states to smooth the pose change. Simultaneously, simulated image acquisition scoring prediction is performed on this transition segment to ensure that the interpolation result does not affect image quality. Under the premise of ensuring that the image projection error does not exceed the maximum tolerance, the interpolation result of this transition segment is used as adjacency correction information when the UAV performs heading registration on the remote sensing mapping path.

[0065] It should be noted that, in this application, the track segment interval refers to the result of dividing the entire flight path of the UAV into bounded segments when performing target mapping; the tracking correction index refers to the set of data features that measure and guide heading adjustments; the pose matching deviation refers to the spatial difference between the actual flight pose of the UAV and the pose required to achieve effective target recognition; the adjacency correction information refers to the set of smooth transition parameters for attitude adjustment between adjacent track segments; and the remote sensing mapping path refers to the flight route of the UAV that sequentially covers the mapping area according to the predetermined flight trajectory when performing remote sensing tasks.

[0066] In this embodiment, extracting the aerial survey viewpoint coordinates of the UAV during aerial survey calibration of the target from the adjacency correction information can be achieved through the following steps:

[0067] Based on the adjacency correction information, a calibration compensation rule is constructed for the UAV to perform aerial survey calibration on the target;

[0068] The calibration compensation rules are mapped to the UAV's onboard camera coordinate system to generate viewpoint attitude features;

[0069] The coordinates of the aerial survey viewpoint when the UAV performs aerial survey calibration on the target are determined based on the viewpoint attitude characteristics.

[0070] In practice, firstly, using adjacency correction information as input, a calibration compensation rule is constructed, incorporating elements such as image offset, viewpoint deviation, and relative position variation. This calibration compensation rule employs a quaternion rotation model combined with Euler transformation to convert the projection differences of the target under different views into attitude adjustment parameters. Then, the calibration compensation rule is input into the UAV control system, and coordinate transformation is performed based on the current UAV attitude calculation data (including the heading angle, pitch angle, and roll angle output by the inertial navigation unit). Through the sensor fusion module of the airborne navigation control unit, the calibration compensation rule is transformed from the geographic coordinate system to the UAV's body coordinate system, and then to the airborne camera coordinate system. During this transformation, the direction cosine matrix can be used to represent the three-dimensional attitude differences as angle adjustment amounts within the camera's field of view, ultimately generating viewpoint attitude features applicable to the target point. Finally, after obtaining each set of viewpoint attitude features, the UAV's flight path is fine-tuned through the real-time flight control system, incorporating several reference sampling points calculated backward from the viewpoint attitude features into the flight path. Each reference sampling point contains three-dimensional spatial coordinates, camera pitch angle, and yaw angle information. By calling the flight path planning module set in the ground station control system and combining it with the digital elevation model, the flight altitude is automatically adjusted, and the location where the UAV should collect images is calculated based on the optimal viewing angle parameters corresponding to each target, ultimately determining the aerial survey viewpoint coordinates of the UAV.

[0071] It should be noted that, in this application, aerial survey calibration refers to the method of calibrating the geometric parameters and spatial positioning of ground targets using sensors mounted on an aircraft; calibration compensation rules refer to the conversion rules for correcting the attitude deviation and spatial error of targets in remote sensing images acquired by UAVs; the airborne camera coordinate system refers to a local spatial rectangular coordinate system established with the center of the lens of the camera mounted on the UAV as the origin; viewpoint attitude characteristics are parameters characterizing the observation angle and relative direction of the UAV camera during the imaging process of the target; and aerial survey viewpoint coordinates refer to the three-dimensional spatial position that the UAV needs to reach to complete attitude imaging of the target when performing aerial survey tasks.

[0072] In step S3, the collaborative planning level of the UAV during active planning of the mapping tilt angle is determined based on the positioning constraint strategy when the UAV locates the target along the flight path and all texture scan marks. The collaborative planning level is then reset in a coordinated manner to obtain the point reset trajectory of the UAV when adjusting the target position in the mapping area.

[0073] In this embodiment, determining the collaborative planning level of the UAV during active planning of the mapping tilt angle, based on the positioning constraint strategy when the UAV locates the target along the flight path and all texture scan markers, can be achieved through the following steps:

[0074] Obtain the positioning constraint strategy when the UAV locates a target along its flight path;

[0075] Determine the planning efficiency boundary of the UAV in active planning of the mapping tilt angle based on all texture scan identifiers;

[0076] The collaborative planning level of the UAV in active planning of surveying tilt angle is determined by the positioning constraint strategy and the planning efficiency boundary.

[0077] In practice, firstly, during the UAV's flight along its flight path, the onboard high-precision GPS receiver and inertial navigation system synchronously record the UAV's position and attitude in real time. A laser rangefinder or structured light ranging module then identifies the spatial position of a standard target deployed on the ground. During this process, by identifying the target's positional error, occlusion, viewing angle deviation, and changes in the survey path's tilt angle, a positioning constraint strategy is constructed, encompassing positional constraints, attitude constraints, and visibility constraints. This strategy serves as the UAV's positioning constraint strategy for target localization along its flight path. Specifically, the effective observation interval along each stage of the flight path can be calculated by combining the positional error model and the constraint cost function. Then, during flight, the onboard imaging system continuously scans the ground texture structure, identifying key visual parameters such as texture complexity, boundary feature density, and texture repetition in different regions, generating texture scan markers. The system utilizes the texture scan identifier to construct a 3D texture perception map and combines it with flight perspective to compare and analyze texture recognition strength, occlusion risk, and edge clarity at different tilt angles. This determines the effectiveness evaluation index for mapping tilt angles in different regions, and the maximum and minimum values ​​of this effective evaluation index are used as the planning efficiency boundary for UAV active mapping tilt angle planning. Finally, the positioning constraint strategy and planning efficiency boundary are input into the active mapping tilt angle planning module, which uses a multi-constraint dynamic path adjustment algorithm to coordinately adjust the UAV's attitude angle and view tilt angle. The multi-constraint dynamic path adjustment algorithm aims to improve texture recognition performance. Under the premise of satisfying positioning constraints, it achieves optimal adaptation of tilt angle continuity and texture response intensity through a dynamic attitude adjustment mechanism. The collaborative planning hierarchy is set according to the principle of hierarchical planning difficulty, mapping different terrain texture complexities and flight constraint strengths to corresponding levels (such as basic layer, intermediate layer, and advanced layer). Each level corresponds to different scheduling priorities and attitude degrees of freedom, thus obtaining the collaborative planning hierarchy for UAV active mapping tilt angle planning.

[0078] It should be noted that, in this application, the positioning constraint strategy refers to the rules that restrict the spatial attitude, flight trajectory, and target recognition angle of the UAV during the execution of the mapping task; the planning efficiency boundary refers to the set of minimum visual recognition accuracy thresholds that need to be met in the active planning of the mapping tilt angle; the collaborative planning level refers to dividing the tilt angle planning task into different complexity levels according to the constraint strength and visual recognition difficulty of the mapping scene; and the active planning of the mapping tilt angle refers to the planning method in which the UAV dynamically adjusts the camera shooting tilt angle according to the terrain and target features to improve the quality of remote sensing data.

[0079] Preferably, in this embodiment, the collaborative planning level is reset in a coordinated manner to obtain the point reset trajectory of the UAV when responding to the adjustment of the target position in the mapping area, with reference to... Figure 3As shown in the figure, this is a flowchart illustrating the process of determining the reset trajectory at each point in some embodiments of this application. In this embodiment, determining the reset trajectory at each point can be achieved using the following steps:

[0080] In step S31, the point control features of the UAV during active planning of the survey tilt angle are extracted from the collaborative planning hierarchy;

[0081] In step S32, the displacement offset trajectory during target position adjustment in the surveying area is determined;

[0082] In step S33, a linkage reset sequence is generated based on the displacement offset trajectory when the UAV responds to the adjustment of the target position in the surveying area;

[0083] In step S34, the point reset trajectory of the UAV when adjusting the target position in the mapping area is determined based on the point control characteristics and the linkage reset sequence.

[0084] In practical implementation, firstly, the set of key control points corresponding to each collaborative planning level is retrieved, i.e., the point-based control features. These features include the spatial location of each control point, its corresponding tilt adjustment range, attitude adjustment parameters, and visual recognition threshold. The UAV control system accesses the planning database and uses these control parameters as the basis for performing surveying tilt adjustment and path reset. The extraction process is based on hierarchical indexing and spatial matching algorithms, filtering high-priority nodes in the collaborative planning levels to ensure that the point-based control features cover key terrain and texture change areas. Next, the actual position changes of the targets are monitored through a real-time ground monitoring system or the UAV's multispectral sensor. The original positioning points and adjusted points of the targets within the surveying area are compared to calculate the three-dimensional displacement offset of each target position. Then, trajectory fitting techniques, such as least squares fitting or spline curve fitting, are used to transform the discrete displacement data of each target into a continuous displacement offset trajectory, i.e., the displacement offset trajectory during target position adjustment in the surveying area. Finally, using the displacement offset trajectory as input, a set of reset action sequences is generated through trajectory segmentation and dynamic programming algorithms. This reset sequence includes attitude adjustment, tilt correction, and spatial position compensation commands required by the UAV at different flight phases. The specific process includes: identifying key turning points in trajectory segmentation; calculating the optimal reset path for each trajectory segment based on the dynamic response capability of the flight control system; arranging all optimal reset paths in chronological order to obtain the linkage reset sequence for the UAV to adjust the target position in the mapping area. Finally, the point-by-point control features are fused with the linkage reset sequence. Specifically, the adjustment points in the linkage reset sequence are matched with the point-by-point control features in the spatial dimension to determine key reset nodes. Then, a three-dimensional path optimization algorithm (such as a gradient descent-based path smoothing algorithm) is used to optimize the flight path between the matching points, ensuring path smoothness and the executability of the reset actions. This generates a continuous point-by-point reset trajectory that meets flight safety standards—the point-by-point reset trajectory for the UAV to adjust the target position in the mapping area.

[0085] It should be noted that in this application, the point-by-point control feature refers to the set of key control point parameters used for specific attitude and tilt adjustment in the UAV mapping path; the displacement offset trajectory refers to the spatial change curve of the actual position of the target in the mapping area relative to the original set position; the linkage reset sequence refers to the set of UAV attitude and path adjustment commands generated based on the target position offset; and the point-by-point reset trajectory refers to the mapping path in which the UAV makes responsive flight adjustments in response to changes in the target position.

[0086] In step S4, quantitative mapping control is performed on the mapping targets within the mapping area based on the aerial survey viewpoint coordinates and the point reset trajectory.

[0087] In this embodiment, quantitative mapping control of the mapping target within the mapping area based on the aerial survey viewpoint coordinates and the sub-point reset trajectory can be achieved through the following steps:

[0088] The attitude contribution of the survey target is determined based on the coordinates of the aerial survey viewpoint.

[0089] The dynamic tracking index of the survey target within the survey area is generated by the point reset trajectory;

[0090] The target mapping control strategy within the mapping area is verified by the attitude contribution and the dynamic tracking index, and adaptive orientation matching is performed.

[0091] In practice, the process begins by acquiring the coordinates of the aerial survey viewpoint. These coordinates are then compared to the preset reference coordinates of each target in the surveying area to calculate the vector angle, yielding the observation angle of each target relative to the current viewpoint. Based on this observation angle, the angle between the illumination direction and the target surface normal, the line-of-sight distance, and imaging distortion parameters, an attitude influence factor model is established. The influence of this model on observation accuracy in three-dimensional space can be calculated using a direction cosine matrix, quantified as an attitude contribution index. This attitude contribution index is represented as a floating-point number between 0 and 1. The closer the attitude contribution of each target is to 1, the higher the imaging accuracy of that aerial survey viewpoint. Then, using the point-reset trajectory as input, the observation points of the mapping target in multiple aerial survey cycles are reconstructed through timestamp calibration and 3D coordinate interpolation algorithms, forming a multi-angle historical observation path for the target. Based on this multi-angle historical observation path, spatial clustering and continuous target visualization analysis are applied, combined with the field of view of the aerial survey platform camera and the gimbal control angle, to statistically analyze the continuous observability, viewing angle change rate, and observation coverage frequency of each target in different trajectory segments; dynamic tracking indicators for the mapping targets within the survey area are generated. Finally, after normalizing the attitude contribution and dynamic tracking indicators corresponding to each target, a binary evaluation vector is constructed, and the comprehensive mapping priority is calculated through a weighted decision function. This weighted decision function can set weight factors through a multi-objective weighted optimization model, assigning different contribution preferences according to different target types (such as point, horizontal, or inclined surfaces). Based on comprehensive priority ranking, an adaptive orientation matching algorithm is used to adjust the strategy matching of the mapping target, including optimizing the image acquisition angle, fine-tuning the flight trajectory, and adjusting the image stitching order. The UAV system iteratively updates the control parameters according to the feedback image clarity and coverage integrity results, so as to realize the closed-loop matching of the mapping process.

[0092] It should be noted that, in this application, attitude contribution refers to the numerical index of the viewpoint's ability to create a stereoscopic imaging effect and high-fidelity reproduction of the surveyed target; dynamic tracking index refers to the ability value of the surveyed target to be continuously observed, identified, and tracked throughout the entire aerial survey path; target surveying control strategy refers to the image acquisition, path selection, and processing optimization scheme adopted for each target in the surveying task; adaptive orientation matching refers to dynamically adjusting the shooting direction and matching parameters according to the external attitude and tracking status during the surveying image acquisition process to improve image fusion quality and surveying model stability.

[0093] Therefore, this application demonstrates a significant improvement in the dynamic recognition and registration capabilities of UAVs for ground-based intelligent targets, addressing the inaccuracy of target coordination responses in existing UAV remote sensing mapping processes. Specifically, by deploying multiple sets of self-identifying intelligent remote sensing targets within the mapping area and assigning thermal tracking and texture scanning markers to each set, dual-channel recognition of the target targets in both thermal infrared and spatial image data is achieved, enhancing the stable tracking performance and recognition accuracy of UAVs in complex terrain. Furthermore, by determining the pose interaction attributes of the UAV while flying along its flight path within the monitoring area and performing segmented registration with all thermal tracking markers, dynamic corrections can be made for errors caused by inaccurate target coordination during flight. To address attitude deviations caused by environmental disturbances, adjacency correction information is generated for heading registration, thereby improving the geometric continuity and registration robustness of remote sensing mapping paths. By determining the collaborative planning hierarchy based on positioning constraint strategies and texture scanning identifiers along the UAV flight path and implementing linked reset, dynamic planning of multi-point flight tilt angles and observation paths can be performed for target positions, improving flight safety and planning efficiency during tilt angle adjustment. By combining aerial survey viewpoint coordinates and point reset trajectories for quantitative mapping control of survey targets, high-precision point reconstruction and dynamic error compensation within the survey area can be achieved, effectively improving the spatial accuracy of remote sensing modeling and the consistency of multi-view fusion.

[0094] In summary, the technical solution adopted in this application can dynamically register the target collaborative control process in UAV remote sensing mapping tasks in an intelligent remote sensing target cluster environment, thereby improving the UAV's mapping response capability to ground targets.

[0095] Example 2: This application provides a collaborative control system for intelligent remote sensing targets based on unmanned aerial vehicles (UAVs), referring to... Figure 4 As shown in the figure, this is a modular structure diagram of a UAV-based intelligent remote sensing target collaborative control system according to this embodiment of the present application. The collaborative control system includes:

[0096] The label setting module 100 is used to set up multiple sets of self-identifying intelligent remote sensing targets in the surveying area, and to set thermal tracking labels and texture scanning labels on each set of targets when the UAV performs remote sensing measurements.

[0097] The segmented registration module 200 is used to determine the pose interaction attributes of the UAV when it flies along the flight path in the monitoring area, and to perform segmented registration of the pose interaction attributes with all thermal tracking identifiers to obtain the adjacency correction information when the UAV performs heading registration on the remote sensing mapping path. Then, the aerial survey viewpoint coordinates of the UAV when performing aerial survey calibration on the target are extracted from the adjacency correction information.

[0098] The linkage reset module 300 is used to determine the collaborative planning level of the UAV during active planning of the mapping tilt angle based on the positioning constraint strategy and all texture scan marks when the UAV locates the target along the flight path. The collaborative planning level is then linked to reset to obtain the point reset trajectory of the UAV when it responds to the adjustment of the target position in the mapping area.

[0099] The quantitative control module 400 is used to perform quantitative mapping control on the mapping targets within the mapping area based on the aerial survey viewpoint coordinates and the point reset trajectory.

[0100] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0101] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, including read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electrically-Erasable Programmable Read-Only Memory (EEPROM), compactdisc read-only memory (CD-ROM) or other optical disc storage, disk storage, magnetic tape storage, or any other computer-readable medium capable of carrying or storing data.

[0102] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

Claims

1. A method for collaborative control of intelligent remote sensing targets based on unmanned aerial vehicles (UAVs), characterized in that, The collaborative control method includes the following steps: Multiple sets of self-identifying intelligent remote sensing targets are deployed in the surveying area, and thermal tracking and texture scanning marks are set on each set of targets for remote sensing measurements by UAVs. Determine the pose interaction attributes of the UAV when it flies along the flight path in the monitoring area, and perform segmented registration of the pose interaction attributes with all thermal tracking markers to obtain the adjacency correction information when the UAV performs heading registration on the remote sensing mapping path. Then, extract the aerial survey viewpoint coordinates of the UAV when performing aerial survey calibration on the target from the adjacency correction information. Specifically, determining the pose and interaction attributes of the UAV while it flies along its flight path within the monitoring area includes: To obtain the pose evolution trend of the UAV as it flies along the flight path; Based on the pose evolution trend, determine the interactive calibration features of the UAV when performing mapping and calibration within the monitoring area; The pose interaction attributes of the UAV when flying along the flight path within the monitoring area are determined based on the interaction calibration features. Specifically, extracting the aerial survey viewpoint coordinates of the UAV during aerial survey calibration of the target from the adjacency correction information includes: Based on the adjacency correction information, a calibration compensation rule is constructed for the UAV to perform aerial survey calibration on the target; The calibration compensation rules are mapped to the UAV's onboard camera coordinate system to generate viewpoint attitude features; The coordinates of the aerial survey viewpoint when the UAV performs aerial survey calibration on the target are determined based on the viewpoint attitude characteristics. Based on the positioning constraint strategy when the UAV locates the target along the flight path and all texture scan marks, the collaborative planning level of the UAV is determined when actively planning the mapping tilt angle. The collaborative planning level is then reset in a coordinated manner to obtain the point reset trajectory of the UAV when it responds to the adjustment of the target position in the mapping area. Specifically, the collaborative planning hierarchy for active planning of the UAV during mapping tilt angle determination, based on the positioning constraint strategy when the UAV locates the target along its flight path and all texture scan markers, includes: Obtain the positioning constraint strategy when the UAV locates a target along its flight path; Determine the planning efficiency boundary of the UAV in active planning of the mapping tilt angle based on all texture scan identifiers; The collaborative planning level of the UAV in active planning of surveying tilt angle is determined by the positioning constraint strategy and the planning efficiency boundary. Quantitative mapping control is performed on the mapping targets within the mapping area based on the aerial survey viewpoint coordinates and the point reset trajectory; Specifically, the quantitative mapping control of the mapping targets within the mapping area based on the aerial survey viewpoint coordinates and the point reset trajectory includes: The attitude contribution of the survey target is determined based on the coordinates of the aerial survey viewpoint. The dynamic tracking index of the survey target within the survey area is generated by the point reset trajectory; The target mapping control strategy within the mapping area is verified by the attitude contribution and the dynamic tracking index, and adaptive orientation matching is performed.

2. The method for collaborative control of intelligent remote sensing targets based on unmanned aerial vehicles (UAVs) as described in claim 1, characterized in that, The aforementioned pose interaction attribute refers to the matching information between the attitude change pattern of the UAV during flight and the target mapping requirements.

3. The method for collaborative control of intelligent remote sensing targets based on unmanned aerial vehicles (UAVs) as described in claim 1, characterized in that, The adjacency correction information refers to the set of smooth transition parameters for attitude adjustment between adjacent track segments.

4. The method for collaborative control of intelligent remote sensing targets based on unmanned aerial vehicles (UAVs) as described in claim 1, characterized in that, The aforementioned active mapping tilt angle planning refers to a planning method for UAVs to dynamically adjust the camera's shooting tilt angle based on terrain and target features in order to improve the quality of remote sensing data.

5. The method for collaborative control of intelligent remote sensing targets based on unmanned aerial vehicles (UAVs) as described in claim 1, characterized in that, The aforementioned positioning constraint strategy refers to the rules that restrict the spatial attitude, flight trajectory, and target recognition angle of the UAV during the execution of surveying and mapping tasks.

6. A UAV-based intelligent remote sensing target cooperative control system, used to execute the UAV-based intelligent remote sensing target cooperative control method as described in any one of claims 1 to 5, characterized in that, The collaborative control system includes: The label setting module is used to deploy multiple sets of self-identifying intelligent remote sensing targets in the surveying area, and to set thermal tracking labels and texture scanning labels on each set of targets when the UAV performs remote sensing measurements; The segmented registration module is used to determine the pose interaction attributes of the UAV when it flies along the flight path in the monitoring area, and to perform segmented registration of the pose interaction attributes with all thermal tracking markers to obtain the adjacency correction information when the UAV performs heading registration on the remote sensing mapping path. Then, the aerial survey viewpoint coordinates of the UAV when performing aerial survey calibration on the target are extracted from the adjacency correction information. The linkage reset module is used to determine the collaborative planning level of the UAV during active planning of the mapping tilt angle based on the positioning constraint strategy and all texture scan marks when the UAV locates the target along the flight path. The collaborative planning level is then linked to reset to obtain the point reset trajectory of the UAV when it responds to the adjustment of the target position in the mapping area. The quantitative control module is used to perform quantitative mapping control on the mapping targets within the mapping area based on the aerial survey viewpoint coordinates and the point reset trajectory.

Images