Air-ground cooperative unmanned aerial vehicle laser wireless energy supply method and system

Abstract

The application provides a kind of air-ground cooperative unmanned aerial vehicle laser wireless power supply method and system, belongs to unmanned aerial vehicle wireless energy transmission technical field.The method comprises: establishing double-link three-dimensional geometric model and evaluating link quality;Through OStrack, target frame and tracking confidence are obtained, and are converted to world coordinate system;Prior prediction is carried out using constant acceleration model, innovation saturation boundary and observation noise covariance are adaptively adjusted according to confidence, posterior motion state is obtained by executing iterative saturated Kalman filtering;Forward compensation is carried out on cumulative system time delay, ground turntable azimuth angle and elevation angle instruction are generated;According to predicted position and speed, working stage is identified, double-source initial power is allocated in combination with stage priority coefficient, link quality and battery demand;Unified safety scaling is carried out according to independent power upper limit of photovoltaic receiver and total power upper limit, final transmission power is outputted and battery energy is updated.The application improves tracking robustness, coverage range and safety of laser power supply in complex environment.

Description

Technical Field

[0001] This invention relates to the field of wireless power transmission technology for unmanned aerial vehicles (UAVs), specifically to a method and system for providing laser wireless power to UAVs in a coordinated air-to-ground manner. Background Technology

[0002] Drones have been widely used in tasks such as power line inspection, disaster monitoring, surveying, logistics, and emergency communications. However, due to limitations in the quality and capacity of onboard batteries, drones have limited continuous operating time, and frequent returns for replacement or recharging reduce mission efficiency. Laser wireless power supply can transmit laser energy from a remote location to the drone's photovoltaic receiver, providing a technological solution to extend drone endurance.

[0003] Existing ground-based laser power supply systems typically rely on fixed ground transmitters. When terrain undulations, buildings, or power transmission facilities obstruct the line-of-sight link from the ground to the UAV, laser power supply is easily interrupted; to cover a large inspection area, multiple ground transmitters are often required. While using only high-altitude platforms for power supply can create a top-down optical path, high-altitude platforms suffer from drift, load, and pointing stability limitations, making it difficult to continuously track high-speed maneuvering UAVs independently. Current technology lacks a mechanism to coordinate the power supply between ground and high-altitude transmitters based on flight phase and link status.

[0004] On the other hand, the narrow laser beam width means that the cumulative delays caused by UAV acceleration, turning, short-term occlusion, and image acquisition, state estimation, communication, and turntable execution all contribute to beam pointing lag. Ordinary visual trackers typically only output the target position, lacking explicit motion state prediction; traditional Kalman filters are also susceptible to abnormal observations caused by occlusion, background interference, atmospheric turbulence, and tracking drift. Directly using unreliable visual observations for state correction amplifies trajectory prediction and beam pointing errors.

[0005] Furthermore, when ground-based laser access points and high-altitude platforms provide power simultaneously, it is necessary to consider the UAV's operational phase, the instantaneous quality of the two links, battery energy requirements, the maximum power of the transmitter, the allowable receiving power of each of the two photovoltaic receivers, and the total received power. Using fixed switching or fixed-ratio allocation can easily lead to power interruptions, forced use of poorer links, or overload of the receiver's optical power.

[0006] Therefore, there is a need for a unified air-to-ground collaborative UAV laser wireless power supply solution that integrates confidence-based tracking and prediction, latency compensation, dual-link evaluation, and receiver security constraints. Summary of the Invention

[0007] The purpose of this invention is to provide a method and system for wireless laser power supply for air-ground cooperative unmanned aerial vehicles (UAVs) to solve at least one technical problem existing in the prior art.

[0008] On one hand, embodiments of the present invention provide a method for air-to-ground cooperative UAV laser wireless power supply. This method is applied to an air-to-ground cooperative power supply system including a ground laser access point, a high-altitude platform, and a UAV equipped with a bottom-mounted photovoltaic receiver and a top-mounted photovoltaic receiver. The method includes: Step S1, acquiring spatial location information of the ground laser access point, the high-altitude platform, the UAV, and a preset hovering point; acquiring the attitude information, battery energy information, and operational status information of the UAV; establishing three-dimensional geometric models from the ground laser access point to the UAV and from the high-altitude platform to the UAV; and calculating the link quality of the two laser power supply links respectively; Step S2, through the ground laser access point... A high-speed camera at the optical access point acquires images of the UAV. The current UAV image and target template are input into a single-stream Transformer visual tracking network to obtain the UAV target bounding box and tracking confidence. The center position of the target bounding box is converted into the UAV position observation in the world coordinate system. Step S3: Establish a constant acceleration motion state model including UAV position, UAV velocity, and UAV acceleration. Calculate the prior state and prior covariance at the current moment based on the posterior state of the previous moment and the process noise covariance. Step S4: Adaptively determine the observation noise covariance and innovation saturation bound based on the tracking confidence. Update the posterior state using iterative saturated Kalman filtering. Step S5: Obtain the posterior motion state of the UAV; Step S6: Obtain the cumulative system delay based on the image acquisition delay, state estimation delay, command transmission delay, and turntable execution delay; Propagate the posterior motion state of the UAV along the constant acceleration motion state model to the moment after the cumulative system delay to obtain the delay compensation position and delay compensation velocity; Calculate the azimuth and pitch commands of the ground laser access point based on the delay compensation position; Step S7: Based on the delay compensation velocity, delay compensation position, tracking confidence, and the distance between the UAV and the next preset hovering point, identify whether the UAV is currently in the flight phase, transition phase, or hovering phase, and calculate the current... The priority coefficients of the ground laser access point and the high-altitude platform corresponding to the previous stage are as follows: Step S7: Calculate the initial laser emission power of the ground laser access point and the high-altitude platform according to the priority coefficients of the ground laser access point and the high-altitude platform, the link quality of the two laser power supply links, and the UAV battery energy information; Step S8: Scale the initial laser emission power according to the maximum allowable receiving power of the bottom photovoltaic receiver, the maximum allowable receiving power of the top photovoltaic receiver, and the maximum allowable total receiving power of the UAV, to obtain the final laser emission power of the ground laser access point and the high-altitude platform and update the UAV battery energy.

[0009] Furthermore, the calculation of the link quality of the two laser power supply links in step S1 includes: the propagation vectors of the ground laser access point and the high-altitude platform relative to the predicted position of the UAV are respectively the ground propagation vectors. and high-altitude propagation vector The corresponding propagation distances are respectively and Calculated as follows: ; ; ; ; In the formula, The predicted location of the UAV after cumulative system delay compensation; Location of the ground laser access point. Location of the high-altitude platform; Link quality is calculated based on the propagation distance of the ground laser access point relative to the predicted position of the UAV, the propagation distance of the high-altitude platform relative to the predicted position of the UAV, the incident factor of the bottom photovoltaic receiver, the incident factor of the top photovoltaic receiver, line-of-sight availability, atmospheric attenuation coefficient, initial spot radius, and beam divergence angle.

[0010] Further, step S2, acquiring UAV images through a high-speed camera set at the ground laser access point, inputting the current UAV image and target template into a single-stream Transformer visual tracking network to obtain the UAV target bounding box and tracking confidence, and converting the center position of the target bounding box into the UAV position observation in the world coordinate system, includes: step S201, dividing the target template in the previous frame or initialization frame and the search area of ​​the current frame into image blocks through OSTrack, performing joint feature interaction through image block embedding, position encoding and single-stream Transformer encoder, and outputting the current target bounding box and tracking confidence; step S202, based on the camera intrinsic parameter matrix and target depth, back-projecting the center of the target bounding box to the camera coordinate system, and then using the rotation matrix and translation vector from the camera to the world coordinate system to obtain the position observation in the world coordinate system.

[0011] Furthermore, step S3, establishing a constant acceleration motion state model including the UAV's position, velocity, and acceleration, and calculating the prior state and prior covariance at the current moment based on the posterior state of the previous moment and the process noise covariance, includes: The state vector of the constant acceleration motion state model is represented as follows: ; In the formula, The three-dimensional position of the drone in motion. The three-dimensional velocity of the drone in motion. The three-dimensional acceleration of the drone in motion; The state transition matrix of the constant acceleration motion state model is represented as follows: ; In the formula, The sampling interval is... It is a 3×3 identity matrix; The mathematical expression for the prior state is: ; In the formula, This is the posterior state of the previous moment. This represents the prior state at the current moment; The mathematical expression for prior covariance is: ; In the formula, The posterior covariance of the previous time step. Let be the process noise covariance.

[0012] Furthermore, in step S4, the observation noise covariance is adaptively determined based on the tracking confidence level. and innovation saturation include: ; ; In the formula, The nominal observation noise covariance, To track confidence levels, To prevent positive numbers from being divided by zero, and These represent the minimum and maximum innovation saturation boundaries, respectively. In step S4, the a posteriori state is updated after iterative saturated Kalman filtering, and the resulting posterior motion state of the UAV includes: ; ; ; ; In the formula, For Kalman filter gain, H represents the iterative innovation quantity, and H is the observation matrix from which the position components are extracted from the state vector. For the observation of the UAV's position, For component-wise saturation operator, As the current posterior state.

[0013] Furthermore, step S6, based on the delay compensation speed, delay compensation position, tracking confidence, and the distance between the UAV and the next preset hovering point, identifies whether the UAV is currently in the flight phase, transition phase, or hovering phase, and calculates the ground laser access point priority coefficient and high-altitude platform priority coefficient corresponding to the current phase, including: Calculate the spatial distance between the time delay compensation position and the next preset hovering point. And calculate the magnitude of the delay compensation speed. ; when Not greater than the preset hovering distance threshold and Not greater than the preset hovering speed threshold When this occurs, it is identified as the hovering phase; when Greater than But not greater than the preset transition distance threshold ,and Greater than But not greater than the preset transition speed threshold At that time, it is identified as a transitional phase; The remaining situations are identified as the flight phase; in, ; Priority coefficient of ground laser access point during flight phase Greater than the priority coefficient of the high-altitude platform ; Priority coefficient of high-altitude platforms during hovering phase Greater than the priority coefficient of ground laser access point ; The dual-source priority factor during the transition phase is based on the progress of the drone approaching hover. The priority coefficients for the flight phase and the hovering phase are obtained by continuous interpolation.

[0014] Furthermore, step S7, calculating the initial laser emission power of the ground laser access point and the high-altitude platform based on the priority coefficient of the ground laser access point, the priority coefficient of the high-altitude platform, the link quality of the two laser power supply links, and the UAV battery energy information at the current stage, includes: The initial laser emission power expression is: ; ; ; In the formula, k is h or g. This represents the initial laser emission power of the high-altitude platform. The initial laser emission power of the ground laser access point. The initial power coefficient, The maximum power of the k-th transmitter. Let be the expected received power of the k-th link. Let K be the optoelectronic link efficiency of the k-th link. For the link quality of the k-th link, It is a constant. The normalized power allocation share, To normalize charging demand, The maximum total required received power is set for the drone.

[0015] Furthermore, step S8, which involves scaling the initial laser emission power proportionally based on the maximum allowable receiving power of the bottom-mounted photovoltaic receiver, the maximum allowable receiving power of the top-mounted photovoltaic receiver, and the maximum allowable total receiving power of the UAV, to obtain the final laser emission power of the ground laser access point and the high-altitude platform and update the UAV battery energy, includes: The final mathematical expression for laser emission power is: ; ; In the formula, k is either g or h. Let the final laser emission power of the k-th link be... To standardize the safety scaling factor, This represents the maximum permissible received power of the bottom-mounted photovoltaic receiver. The maximum allowable receiving power for a rooftop photovoltaic receiver is, This represents the maximum permissible total received power for the drone. After photovoltaic conversion and charging circuitry, the drone's battery energy is updated according to the following formula: ; In the formula, For photoelectric conversion efficiency, For charging efficiency, This represents the total power consumption of the drone at its current stage. To control the cycle, This represents the current battery power of the drone.

[0016] Furthermore, after step S4, the method further includes: When the tracking confidence is lower than a preset low confidence threshold, skip the observation correction and output the prior state; when the number of consecutive frames in which the tracking confidence is lower than the preset low confidence threshold reaches a preset re-detection threshold, expand the search area or reinitialize the target template.

[0017] Secondly, embodiments of the present invention also provide an air-to-ground cooperative UAV laser wireless power supply system. The system is implemented using the aforementioned air-to-ground cooperative UAV laser wireless power supply method. The system includes: a status acquisition module, adapted to acquire spatial position information of the ground laser access point, the high-altitude platform, the UAV, and a preset hovering point, and to acquire the attitude information, battery energy information, and operational status information of the UAV; a geometric modeling and link evaluation module, adapted to establish three-dimensional geometric models of the ground laser access point to the UAV and the high-altitude platform to the UAV, and to calculate the link quality of the two laser power supply links respectively; and a visual tracking and coordinate transformation module, adapted to track the UAV through a high-altitude platform set at the ground laser access point. A high-speed camera acquires images of the UAV, and the current UAV image and target template are input into a single-stream Transformer visual tracking network to obtain the UAV target bounding box and tracking confidence. The center position of the target bounding box is converted into the UAV position observation in the world coordinate system. The confidence-aware state estimation module is suitable for establishing a constant acceleration motion state model that includes the UAV position, UAV velocity, and UAV acceleration. It calculates the prior state and prior covariance at the current moment based on the posterior state at the previous moment and the process noise covariance. Based on the tracking confidence, the observation noise covariance and innovation saturation bound are adaptively determined. After iterative saturated Kalman filtering, the posterior state is updated to obtain the UAV posterior motion state. The delay compensation and beam pointing module is suitable for obtaining the cumulative system delay based on image acquisition delay, state estimation delay, command transmission delay, and turntable execution delay. It then propagates the UAV's posterior motion state forward along a constant acceleration motion state model to the moment after the cumulative system delay, obtaining the delay compensation position and delay compensation velocity. Based on the delay compensation position, it calculates the azimuth and pitch commands of the ground laser access point. The operation phase identification module is suitable for identifying whether the UAV is currently in a flight phase, transition phase, or hovering phase based on the delay compensation velocity, delay compensation position, tracking confidence, and the distance between the UAV and the next preset hovering point, and calculates the corresponding ground laser access point for the current phase. The system includes a ground laser access point priority coefficient and an aerial platform priority coefficient; a dual-source power allocation module, suitable for calculating the initial laser emission power of the ground laser access point and the aerial platform based on the current stage's ground laser access point priority coefficient, aerial platform priority coefficient, link quality of the two laser power supply links, and UAV battery energy information; and a safety scaling and battery update module, suitable for scaling the initial laser emission power proportionally based on the maximum allowable receiving power of the bottom photovoltaic receiver, the maximum allowable receiving power of the top photovoltaic receiver, and the maximum allowable total receiving power of the UAV, to obtain the final laser emission power of the ground laser access point and the aerial platform and update the UAV battery energy.

[0018] Thirdly, embodiments of the present invention also provide an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the above-described air-to-ground cooperative UAV laser wireless power supply method.

[0019] Fourthly, embodiments of the present invention also provide a readable storage medium, which, when the instructions in the storage medium are executed by the processor of an electronic device, enables the electronic device to execute the above-described air-to-ground cooperative UAV laser wireless power supply method.

[0020] The beneficial effects of this invention are as follows: 1. By having ground-based laser access points and high-altitude platforms work together to provide power, the continuity of power supply can be improved when ground links are blocked, UAVs are outside the effective range of the ground, or UAVs are in the hovering phase, and the need for densely deploying ground-based transmitters for large-scale missions can be reduced.

[0021] 2. By receiving downward and upward laser beams through top-mounted and bottom-mounted photovoltaic receivers respectively, the geometric obstruction of the UAV body to the optical path in opposite directions is reduced, and the two optical links are allowed to work simultaneously or in coordination.

[0022] 3. By using the tracking confidence output of OSTrack to jointly adjust the observation noise covariance and innovation saturation bound, the impact of abnormal observations caused by occlusion, turbulence, background interference and tracking drift on motion state estimation can be limited; the pure prediction and re-detection mechanism can maintain the state output during short-term target loss.

[0023] 4. By forward compensation of the cumulative delay generated by image acquisition, state estimation, command transmission and turntable execution, and by converting the future position into azimuth and pitch commands, the lag of the laser beam in high-speed motion can be reduced.

[0024] 5. By using the operational phase, link quality, battery requirements, and transmitter capabilities together for initial power allocation, and then satisfying the independent upper limits and total received power upper limits of the two receivers in a uniform ratio, receiver overload can be avoided while maintaining the dual-source allocation ratio. Attached Figure Description

[0025] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0026] Figure 1 This is a schematic diagram of the air-to-ground collaborative UAV laser wireless power supply method provided in Embodiment 1 of the present invention.

[0027] Figure 2 This is a schematic diagram of the overall architecture of an air-ground collaborative UAV laser wireless power supply system provided in Embodiment 1 of the present invention.

[0028] Figure 3 This is a schematic diagram of the structure and light-receiving direction of a top-mounted and bottom-mounted photovoltaic receiver for a drone provided in Embodiment 1 of the present invention.

[0029] Figure 4 This is a schematic diagram of a three-dimensional geometric model of a laser wireless power supply system for an air-to-ground collaborative UAV provided in Embodiment 1 of the present invention.

[0030] Figure 5 This is a schematic diagram of a laser wireless power supply system for an air-to-ground collaborative UAV provided in Embodiment 2 of the present invention.

[0031] Figure 6 This is a partial block diagram of the electronic device provided in Embodiment 3 of the present invention. Detailed Implementation

[0032] Before discussing the exemplary embodiments in more detail, it should be noted that some exemplary embodiments are described as processes or methods depicted as flowcharts. Although the flowcharts describe the operations as sequential processes, many of these operations can be performed in parallel, concurrently, or simultaneously. Furthermore, the order of the operations can be rearranged. The process can be terminated when its operation is completed, but may also have additional steps not included in the figures. The process can correspond to a method, function, procedure, subroutine, subroutine, etc.

[0033] It should be understood that although the terms "first," "second," etc., may be used herein to describe various units, these units should not be limited by these terms. These terms are used merely to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, a first unit may be referred to as a second unit, and similarly, a second unit may be referred to as a first unit. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0034] The present invention will now be described in detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, illustrating only the basic structure of the invention, and therefore only show the components relevant to the invention.

[0035] For ease of understanding, the following technical terms are explained here: HAP: High Altitude Platform; UAV: Unmanned Aerial Vehicle; LAP: Laser Access Point.

[0036] Example 1 The specific implementation method is as follows: like Figure 1 The diagram shown is a flowchart of a laser wireless power supply method for air-to-ground cooperative UAVs provided by the present invention.

[0037] As an example, combined Figure 2 and Figure 3As shown, the method is applied to an air-ground coordinated power supply system including a ground laser access point (LAP), a high-altitude platform (HAP), and a drone (UAV) equipped with a bottom-mounted photovoltaic receiver and a top-mounted photovoltaic receiver. The method includes: Step S1, acquiring the spatial location information of the ground laser access point, the high-altitude platform, the drone, and a preset hovering point; acquiring the attitude information, battery energy information, and operational status information of the drone; establishing three-dimensional geometric models of the ground laser access point to the drone and the high-altitude platform to the drone; and calculating the link quality of the two laser power supply links respectively; Step S2, using a high-speed camera installed at the ground laser access point to collect data. Collect UAV images, input the current UAV image and target template into a single-stream Transformer visual tracking network to obtain the UAV target bounding box and tracking confidence, and convert the center position of the target bounding box into the UAV position observation in the world coordinate system; Step S3: Establish a constant acceleration motion state model including UAV position, UAV velocity, and UAV acceleration, and calculate the prior state and prior covariance at the current moment based on the posterior state at the previous moment and the process noise covariance; Step S4: Adaptively determine the observation noise covariance and innovation saturation bound based on the tracking confidence, and update the posterior state using iterative saturated Kalman filtering to obtain the UAV posterior state. Step S5: Based on the image acquisition delay, state estimation delay, command transmission delay, and turntable execution delay, obtain the cumulative system delay. Propagate the UAV's posterior motion state forward along the constant acceleration motion state model to the moment after the cumulative system delay to obtain the delay compensation position and delay compensation velocity. Calculate the azimuth and pitch commands of the ground laser access point based on the delay compensation position. Step S6: Based on the delay compensation velocity, delay compensation position, tracking confidence, and the distance between the UAV and the next preset hovering point, identify whether the UAV is currently in the flight phase, transition phase, or hovering phase, and calculate the impact of the current phase on the UAV's position. The corresponding ground laser access point priority coefficient and high-altitude platform priority coefficient are determined; Step S7: Based on the ground laser access point priority coefficient, high-altitude platform priority coefficient, link quality of the two laser power supply links, and UAV battery energy information, the initial laser emission power of the ground laser access point and high-altitude platform are calculated respectively; Step S8: Based on the maximum allowable receiving power of the bottom photovoltaic receiver, the maximum allowable receiving power of the top photovoltaic receiver, and the maximum allowable total receiving power of the UAV, the initial laser emission power is scaled by a uniform ratio to obtain the final laser emission power of the ground laser access point and high-altitude platform and update the UAV battery energy.

[0038] In some feasible implementations, the calculation of the link quality of the two laser power supply links in step S1 includes: the propagation vectors of the ground laser access point and the high-altitude platform relative to the predicted position of the UAV are respectively the ground propagation vectors. and high-altitude propagation vector The corresponding propagation distances are respectively and Calculated as follows: ; ; ; ; In the formula, The predicted location of the UAV after cumulative system delay compensation; Location of the ground laser access point. Location of the high-altitude platform; Link quality is calculated based on the propagation distance of the ground laser access point relative to the predicted position of the UAV, the propagation distance of the high-altitude platform relative to the predicted position of the UAV, the incident factor of the bottom photovoltaic receiver, the incident factor of the top photovoltaic receiver, line-of-sight availability, atmospheric attenuation coefficient, initial spot radius, and beam divergence angle.

[0039] Preferred, combined Figure 4 As shown, in the global coordinate system, the position of the nth ground laser access point is denoted as... After selecting the currently serving terrestrial laser access point, use that terrestrial laser access point as the local coordinate origin to obtain the location of the terrestrial laser access point. Location of the high-altitude platform drone location For the j-th task point that requires hovering to complete data collection or operations, its position is denoted as . .

[0040] Since the subsequent laser pointing does not directly use the current observation position which has processing delay, but uses the delay-compensated position obtained in step S5, the propagation vectors and propagation distances of the ground and high-altitude links are calculated according to the following formula: ; ; ; .

[0041] Using the downward and upward unit vectors in the body coordinate system as references, the external normal vectors of the bottom-mounted and top-mounted photovoltaic receivers in the world coordinate system are obtained, respectively. The dot product of the receiver's external normal vector and the unit vector pointing from the UAV to the corresponding transmitter is then taken, and negative values ​​are truncated to zero to obtain the incident factor of the bottom-mounted photovoltaic receiver. and incident factor of rooftop photovoltaic receiver Let the obstacle region be... When the line segment from the transmitter to the predicted drone position intersects with... When they intersect, the corresponding line-of-sight availability Set the value to 0, otherwise set it to 1. Link quality is calculated using the following formula: ; In the formula, k takes the form of g or h, representing ground link or high-altitude link respectively; Atmospheric attenuation coefficient, The beam divergence angle, This is a constant determined by the initial spot radius and the propagation model. The link quality takes into account distance attenuation, beam spread, incident direction at the receiver surface, and line-of-sight obstruction.

[0042] In some feasible implementations, step S2, acquiring UAV images through a high-speed camera set at the ground laser access point, inputting the current UAV image and target template into a single-stream Transformer visual tracking network to obtain the UAV target bounding box and tracking confidence, and converting the center position of the target bounding box into a UAV position observation in the world coordinate system, includes: step S201, dividing the target template in the previous frame or initialization frame and the search area of ​​the current frame into image blocks through OSTrack, performing joint feature interaction through image block embedding, position encoding and single-stream Transformer encoder, and outputting the current target bounding box and tracking confidence; step S202, based on the camera intrinsic parameter matrix and target depth, back-projecting the center of the target bounding box to the camera coordinate system, and then using the rotation matrix and translation vector from the camera to the world coordinate system to obtain the position observation in the world coordinate system.

[0043] Preferably, a high-speed camera at the ground-based laser access point continuously acquires images of the UAV. OSTrack (One-Stream Tracking, a visual target tracking framework) divides the target template from the previous or initial frame and the search region of the current frame into image blocks. These blocks undergo joint feature interaction through image block embedding, positional encoding, and a single-stream Transformer encoder to output the current target bounding box. and tracking confidence The center pixel coordinates of the target bounding box are... , and These represent the width and height of the bounding box, respectively. Tracking confidence. The value ranges from 0 to 1 and is used to describe the reliability of the current visual positioning result.

[0044] Based on camera intrinsic parameter matrix and target depth The center of the target bounding box is back-projected onto the camera coordinate system, and then the rotation matrix from the camera to the world coordinate system is used. Translation vector Obtain the position observation in the world coordinate system : ; ; In the formula, Let (x, y, z) be the three-dimensional coordinates of the UAV in the camera coordinate system. The inverse of the camera intrinsic parameter matrix. Let be the rotation matrix from the camera coordinate system to the world coordinate system. This is the translation vector from the camera coordinate system to the world coordinate system. Target depth. Depth can be obtained through laser ranging, binocular vision, structured light, millimeter-wave ranging, or calibrated monocular depth estimation. Camera extrinsic parameters are obtained through calibration after installation.

[0045] In some feasible implementations, step S3 includes: The state vector of the constant acceleration motion model is represented as follows: ; In the formula, The three-dimensional position of the drone in motion. The three-dimensional velocity of the drone in motion. The three-dimensional acceleration of the drone in motion; The state transition matrix of the constant acceleration motion model is represented as follows: ; In the formula, The sampling interval is... It is a 3×3 identity matrix; The mathematical expression for the prior state is: ; In the formula, This is the posterior state of the previous moment. This represents the prior state at the current moment; The mathematical expression for prior covariance is: ; In the formula, The posterior covariance of the previous time step. Let be the process noise covariance.

[0046] In some feasible implementations, step S4, adaptively determining the observation noise covariance and innovation saturation bound based on the tracking confidence, and performing iterative saturated Kalman filtering to update the a posteriori state and obtain the UAV's posterior motion state, includes: based on the tracking confidence output by OSTrack. Observation noise covariance and innovation saturation Adjust according to the following formula: ; ; In the formula, The nominal observation noise covariance, To track confidence levels, To prevent positive numbers from being divided by zero, and These represent the minimum and maximum innovation saturation boundaries, respectively. When the confidence level decreases, Increasing the weight of current observations in the fusion reduces their weight; at the same time This reduces the likelihood of abnormal innovations being truncated earlier.

[0047] The observation matrix H is used to extract the position component from the state vector. Kalman gain, iterative innovation, saturation innovation, and state correction are calculated according to the following formula: ; ; ; ; Saturation operator Each component of the innovation vector is processed separately: when the innovation component is between the corresponding positive and negative saturation boundaries, its original value is retained; when it exceeds the positive saturation boundary, the positive saturation boundary is applied; and when it is below the negative saturation boundary, the negative saturation boundary is applied. This component-by-component saturation processing effectively suppresses the contamination of state estimation by abnormal observations generated by visual tracking in challenging scenarios such as complex backgrounds and pose changes. After J iterations, the state of the Jth iteration is taken as the current posterior state. The posterior covariance can be updated using the Joseph form to maintain the positive semidefiniteness of the covariance matrix.

[0048] Preferably, after step S4, the method further includes: when the tracking confidence is lower than a preset low confidence threshold, skipping observation correction and outputting the prior state; when the number of consecutive frames with tracking confidence lower than the preset low confidence threshold reaches a preset re-detection threshold, expanding the search area or re-initializing the target template. That is, when Below the low confidence threshold At this point, visual observations are considered unreliable, so measurement corrections are skipped, and the posterior state is set equal to the prior state. The number of consecutive low-confidence frames is recorded. ,when Not less than the re-detection threshold If necessary, the OSTrack search area is expanded or the target template is reinitialized; after the target is recovered, the confidence-aware state update is re-executed. Regardless of whether it is a normal update or pure prediction, the current state is fed back to the image plane to update the center and scale of the search area in the next frame.

[0049] In some feasible implementations, step S5, obtaining the cumulative system delay based on image acquisition delay, state estimation delay, command transmission delay, and turntable execution delay, propagating the UAV's posterior motion state forward along the constant acceleration motion state model to the time after the cumulative system delay, obtaining the delay compensation position and delay compensation velocity, and calculating the azimuth and pitch commands of the ground laser access point based on the delay compensation position, includes: Image acquisition, image transmission, visual tracking, state estimation, control command transmission, and ATP turntable execution all incur time delays. The cumulative system delay is obtained by adding the delays of each component. The posterior state is propagated forward along the constant acceleration model to... Extract the future location and calculate the azimuth and elevation angles of the ground laser access point: ; ; ; In the formula, To make the current posterior state Forward propagation Predicted state in seconds, Here is the state transition matrix. This is the azimuth angle at which the ground turntable needs to rotate. This is the pitch angle at which the ground turntable needs to rotate. , , These are the three axis components of the ground pointing vector in the world coordinate system.

[0050] That is, when the beam is not compensated, it points to the current posterior position. In reality, the UAV continues to move within the cumulative time delay, causing the beam to lag behind. After adopting forward prediction, the ATP turntable points to the future position, making the center of the laser beam closer to the expected position of the bottom photovoltaic receiver.

[0051] In some feasible implementations, step S6, identifying whether the UAV is currently in a flight phase, transition phase, or hovering phase based on the delay compensation speed, delay compensation position, tracking confidence, and the distance between the UAV and the next preset hovering point, and calculating the ground laser access point priority coefficient and high-altitude platform priority coefficient corresponding to the current phase, includes: calculating the spatial distance between the delay compensation position and the next preset hovering point. And calculate the magnitude of the delay compensation speed. ;when Not greater than the preset hovering distance threshold and Not greater than the preset hovering speed threshold When, it is identified as the hovering phase; when Greater than But not greater than the preset transition distance threshold ,and Greater than But not greater than the preset transition speed threshold When the transition phase occurs, it is identified as a transition phase; otherwise, it is identified as a flight phase. Priority coefficient of ground laser access point during flight phase Greater than the priority coefficient of the high-altitude platform Priority coefficient for high-altitude platforms during hovering phase Greater than the priority coefficient of ground laser access point The dual-source priority coefficient during the transition phase is based on the progress of the drone approaching hover. The priority coefficients for the flight phase and the hovering phase are obtained by continuous interpolation.

[0052] Preferably, the requirements for the launch source are completely different depending on whether the drone is passing by (flying), preparing to land (transition), or hovering (stationary operation). Therefore, let... To predict the position to the next hovering point distance, The modulus of the latency compensation speed. Set the hovering distance threshold. Hovering speed threshold Transition distance threshold and transition speed threshold and satisfy The running phase m(t) and the transition period. Determined according to the following formula: .

[0053] F, T, and H represent the flight, transition, and hovering phases, respectively. This is the confidence threshold for phase switching. The transition phase is based on... Linear interpolation between the flight coefficient and hovering coefficient allows for a smooth transition of the main power supply role.

[0054] In some feasible implementations, step S7, calculating the initial laser emission power of the ground laser access point and the high-altitude platform based on the priority coefficient of the ground laser access point, the priority coefficient of the high-altitude platform, the link quality of the two laser power supply links, and the UAV battery energy information, includes: The initial laser emission power expression is: ; ; ; In the formula, k is h or g. This represents the initial laser emission power of the high-altitude platform. The initial laser emission power of the ground laser access point. The initial power coefficient, The maximum power of the k-th transmitter. Let be the expected received power of the k-th link. Let K be the optoelectronic link efficiency of the k-th link. For the link quality of the k-th link, It is a constant. The normalized power allocation share, To normalize charging demand, The maximum total required received power is set for the drone.

[0055] Preferably, based on the current battery energy E(t) and battery capacity of the drone Calculate normalized charging demand Multiplying the phase priority coefficient by the link quality yields the unnormalized weight, which is then normalized to obtain the power share of the ground link and the high-altitude link. ; ; ; In the formula, k is the link index, representing g (ground link) or h (high-altitude link). For unnormalized weights, This is the stage priority coefficient. For link quality, This represents the normalized power share.

[0056] When both links are unavailable Both transmitters stop laser output; when at least one link is available, .

[0057] Projection operator The initial power factor is limited to between 0 and 1 to ensure that the initial transmit power does not exceed the hardware capabilities of the ground-based laser access point or the high-altitude platform. The phase priority factor provides a preference rather than a rigid selection; when the nominal primary link quality is severely degraded or blocked, another link can automatically receive a higher share.

[0058] In some feasible implementations, step S8, which involves scaling the initial laser emission power proportionally based on the maximum allowable receiving power of the bottom-mounted photovoltaic receiver, the maximum allowable receiving power of the top-mounted photovoltaic receiver, and the maximum allowable total receiving power of the UAV, to obtain the final laser emission power of the ground laser access point and the high-altitude platform and update the UAV battery energy, includes: The final mathematical expression for laser emission power is: ; ; In the formula, k is either g or h. Let the final laser emission power of the k-th link be... To standardize the safety scaling factor, This represents the maximum permissible received power of the bottom-mounted photovoltaic receiver. The maximum allowable receiving power for a rooftop photovoltaic receiver is, This represents the maximum permissible total received power for the drone. After photovoltaic conversion and charging circuitry, the drone's battery energy is updated according to the following formula: ; In the formula, For photoelectric conversion efficiency, For charging efficiency, This represents the total power consumption of the drone at its current stage. To control the cycle, This represents the current battery power of the drone.

[0059] Preferably, the power consumption during the flight phase can be set to the propulsion system power consumption, the transition phase uses the transition power consumption, and the hovering phase uses the sum of the hovering power consumption and the mission payload power consumption. The projection operator limits the updated battery energy to 0 to... between.

[0060] In some feasible implementations, among a set of optional implementation parameters, the effective area of ​​the bottom-mounted photovoltaic receiver is 0.012 m², and the effective area of ​​the top-mounted photovoltaic receiver is 0.010 m²; the maximum allowable received optical power of both photovoltaic receivers is 120 W, and the maximum allowable total received optical power of the UAV is 200 W. The control cycle can be set to 0.02 s, the low confidence threshold can be set to 0.3, re-detection is triggered when there are 5 consecutive frames of low confidence, and the cumulative system latency can be measured and updated within the range of 0 to 150 ms according to the actual hardware. The above values ​​are only implementation examples and can be adjusted according to the photovoltaic devices, laser wavelength, heat dissipation capacity, mission payload, and flight platform.

[0061] When the UAV is flying at high speed and the ground link is unobstructed, the ground laser access point undertakes the main power supply through real-time visual tracking, state estimation and latency compensation, while the high-altitude platform provides supplementary power. When the ground link is blocked, its link quality drops to zero, and the high-altitude platform undertakes the available power supply. When the UAV approaches the hovering point, the transition progress causes the ground power to gradually decrease and the high-altitude power to gradually increase. During the hovering phase, the high-altitude platform forms a relatively stable top-down main power supply link, and the ground link can provide auxiliary power when available.

[0062] In summary, this invention achieves full-airspace coverage and link redundancy for laser power supply through a dual-source collaborative architecture of a ground-based laser access point and a high-altitude platform, combined with a top-and-bottom photovoltaic receiver design on the UAV. By using the OSTrack visual tracking network and iterative saturated Kalman filtering driven by tracking confidence, it effectively suppresses the contamination of motion state estimation by abnormal observations in complex environments, improving the robustness of tracking and pointing. Forward compensation of the cumulative system delay generated in image acquisition, state estimation, command transmission, and turntable execution significantly reduces beam pointing lag in high-speed maneuvering scenarios. Through a dual-source power allocation mechanism based on operational awareness and a unified safety scaling strategy at the receiver, it meets the UAV's charging needs while considering link quality differences, transmitter hardware constraints, and the independent and total power limits of the receiver's photovoltaic cells, achieving a balance between power supply efficiency and safety. This invention provides a complete technical solution for wireless laser power supply for UAVs, encompassing perception, prediction, pointing, and power management, and can be widely applied in scenarios requiring long-endurance UAV operations, such as power line inspection, emergency communication, and disaster monitoring.

[0063] Example 2 Please see Figure 5 This embodiment provides a schematic diagram of a laser wireless power supply system for air-to-ground collaborative unmanned aerial vehicles.

[0064] As an example, the system is implemented using the air-to-ground cooperative UAV laser wireless power supply method described in Embodiment 1, and the system includes: Status acquisition module 1 is suitable for acquiring the spatial position information of the ground laser access point, the high-altitude platform, the UAV and the preset hovering point, and acquiring the attitude information, battery energy information and operating status information of the UAV; The geometric modeling and link evaluation module 2 is suitable for establishing three-dimensional geometric models of the ground laser access point to the UAV and the high-altitude platform to the UAV, and calculating the link quality of the two laser power supply links respectively. The visual tracking and coordinate transformation module 3 is suitable for acquiring UAV images through a high-speed camera set at the ground laser access point, inputting the current UAV image and target template into a single-stream Transformer visual tracking network to obtain the UAV target bounding box and tracking confidence, and converting the center position of the target bounding box into the UAV position observation in the world coordinate system. The confidence perception state estimation module 4 is suitable for establishing a constant acceleration motion state model that includes the UAV position, UAV speed and UAV acceleration. It calculates the prior state and prior covariance at the current moment based on the posterior state at the previous moment and the process noise covariance. It adaptively determines the observation noise covariance and innovation saturation bound based on the tracking confidence, and updates the posterior state after iterative saturated Kalman filtering to obtain the posterior motion state of the UAV. The delay compensation and beam pointing module 5 is suitable for obtaining the cumulative system delay based on the image acquisition delay, state estimation delay, command transmission delay and turntable execution delay, propagating the UAV's posterior motion state forward along the constant acceleration motion state model to the moment after the cumulative system delay, obtaining the delay compensation position and delay compensation speed, and calculating the azimuth and pitch commands of the ground laser access point based on the delay compensation position; The operation phase identification module 6 is suitable for identifying whether the UAV is currently in the flight phase, transition phase, or hovering phase based on the time delay compensation speed, time delay compensation position, tracking confidence, and the distance between the UAV and the next preset hovering point, and calculating the ground laser access point priority coefficient and high-altitude platform priority coefficient corresponding to the current phase. The dual-source power allocation module 7 is suitable for calculating the initial laser emission power of the ground laser access point and the high-altitude platform based on the priority coefficient of the ground laser access point, the priority coefficient of the high-altitude platform, the link quality of the two laser power supply links, and the battery energy information of the UAV at the current stage.

[0065] The safety scaling and battery update module 8 is suitable for scaling the initial laser emission power proportionally based on the maximum allowable receiving power of the bottom photovoltaic receiver, the maximum allowable receiving power of the top photovoltaic receiver, and the maximum allowable total receiving power of the UAV, so as to update the UAV battery energy by obtaining the final laser emission power of the ground laser access point and the high-altitude platform.

[0066] It is not difficult to see that this embodiment is a system implementation corresponding to the first embodiment, and this embodiment can be implemented in conjunction with the first embodiment. The relevant technical details mentioned in the first embodiment are still valid in this embodiment, and will not be repeated here to reduce repetition. Accordingly, the relevant technical details mentioned in this embodiment can also be applied to the first embodiment.

[0067] It is worth mentioning that all modules involved in this embodiment are logical units. In practical applications, a logical unit can be a physical unit, a part of a physical unit, or a combination of multiple physical units. Furthermore, to highlight the innovative aspects of this invention, this embodiment does not introduce units that are not closely related to solving the technical problem proposed by this invention; however, this does not mean that other units are absent from this embodiment.

[0068] Example 3 Please see Figure 6 The present invention also provides an electronic device, including: a memory and a processor; the memory stores at least one program instruction; the processor loads and executes the at least one program instruction to implement the air-to-ground cooperative UAV laser wireless power supply method provided in Embodiment 1.

[0069] The memory 702 and processor 701 are connected via a bus, which may include any number of interconnecting buses and bridges, connecting various circuits of one or more processors 701 and memory 702 together. The bus may also connect various other circuits, such as peripheral devices, voltage regulators, and power management circuits, which are well known in the art and therefore will not be described further herein. A bus interface provides an interface between the bus and the transceiver. The transceiver may be a single element or multiple elements, such as multiple receivers and transmitters, providing a unit for communicating with various other devices over a transmission medium. Data processed by processor 701 is transmitted over a wireless medium via an antenna, which further receives data and transmits it to processor 701.

[0070] Processor 701 is responsible for managing the bus and general processing, and can also provide various functions, including timing, peripheral interfaces, voltage regulation, power management, and other control functions. Memory 702 can be used to store data used by processor 701 during operation.

[0071] Example 4 This invention also proposes a storage medium storing a method for wireless laser power supply for air-to-ground cooperative UAVs. When the air-to-ground cooperative UAV laser wireless power supply program is executed by a processor, it implements the steps of the method described above. Since this storage medium employs all the technical solutions of the above embodiments, it possesses at least all the beneficial effects brought about by the technical solutions of the above embodiments, which will not be elaborated further here.

[0072] The above descriptions are merely embodiments of the present invention. Commonly known structures and characteristics are not described in detail here. Those skilled in the art are aware of all common technical knowledge in the field prior to the application date or priority date, are aware of all existing technologies in that field, and have the ability to apply conventional experimental methods prior to that date. Those skilled in the art can, based on the guidance provided in this application, improve and implement this solution in combination with their own capabilities. Some typical known structures or methods should not be obstacles for those skilled in the art to implement this application. It should be noted that those skilled in the art can make several modifications and improvements without departing from the structure of the present invention. These should also be considered within the scope of protection of the present invention, and will not affect the effectiveness of the implementation of the present invention or the practicality of the patent. The scope of protection claimed in this application should be determined by the content of its claims, and the specific embodiments described in the specification can be used to interpret the content of the claims.

Claims

1. A method for wireless laser power supply for air-to-ground cooperative unmanned aerial vehicles, characterized in that, The method is applied to an air-to-ground coordinated power supply system including a ground laser access point, a high-altitude platform, and a UAV equipped with a bottom-mounted photovoltaic receiver and a top-mounted photovoltaic receiver. The method includes: Step S1: Obtain the spatial location information of the ground laser access point, the high-altitude platform, the UAV and the preset hovering point; obtain the attitude information, battery energy information and operating status information of the UAV; establish a three-dimensional geometric model from the ground laser access point to the UAV and from the high-altitude platform to the UAV; and calculate the link quality of the two laser power supply links respectively. Step S2: Acquire UAV images using a high-speed camera set at the ground laser access point, input the current UAV image and target template into a single-stream Transformer visual tracking network to obtain the UAV target bounding box and tracking confidence, and convert the center position of the target bounding box into the UAV position observation in the world coordinate system; Step S3: Establish a constant acceleration motion state model that includes the drone's position, speed, and acceleration. Calculate the prior state and prior covariance at the current moment based on the posterior state of the previous moment and the process noise covariance. Step S4: Based on the tracking confidence, adaptively determine the observation noise covariance and innovation saturation bound, and use iterative saturated Kalman filtering to complete the a posteriori state update to obtain the UAV's a posteriori motion state; Step S5: Obtain the cumulative system delay based on the image acquisition delay, state estimation delay, command transmission delay, and turntable execution delay. Propagate the UAV's posterior motion state forward along the constant acceleration motion state model to the time after the cumulative system delay to obtain the delay compensation position and delay compensation velocity. Calculate the azimuth and pitch commands of the ground laser access point based on the delay compensation position. Step S6: Based on the time delay compensation speed, time delay compensation position, tracking confidence and the distance between the UAV and the next preset hovering point, identify whether the UAV is currently in the flight phase, transition phase or hovering phase, and calculate the ground laser access point priority coefficient and high-altitude platform priority coefficient corresponding to the current phase. Step S7: Calculate the initial laser emission power of the ground laser access point and the high-altitude platform based on the priority coefficient of the ground laser access point, the priority coefficient of the high-altitude platform, the link quality of the two laser power supply links, and the battery energy information of the UAV at the current stage. Step S8: Based on the maximum allowable receiving power of the bottom photovoltaic receiver, the maximum allowable receiving power of the top photovoltaic receiver, and the maximum allowable total receiving power of the UAV, the initial laser emission power is scaled up by a uniform ratio to obtain the final laser emission power of the ground laser access point and the high-altitude platform, and the UAV battery energy is updated.

2. The air-to-ground cooperative UAV laser wireless power supply method according to claim 1, characterized in that, The calculation of the link quality of the two laser power supply links in step S1 includes: The propagation vectors of the ground laser access point and the high-altitude platform relative to the predicted position of the UAV are respectively the ground propagation vector. and high-altitude propagation vector The corresponding propagation distances are respectively and Calculated as follows: ； ； ； ； In the formula, The predicted location of the UAV after cumulative system delay compensation; Location of the ground laser access point. Location of the high-altitude platform; Link quality is calculated based on the propagation distance of the ground laser access point relative to the predicted position of the UAV, the propagation distance of the high-altitude platform relative to the predicted position of the UAV, the incident factor of the bottom photovoltaic receiver, the incident factor of the top photovoltaic receiver, line-of-sight availability, atmospheric attenuation coefficient, initial spot radius, and beam divergence angle.

3. The air-to-ground cooperative UAV laser wireless power supply method according to claim 1, characterized in that, Step S2, which involves acquiring UAV images using a high-speed camera located at the ground laser access point, inputting the current UAV image and target template into a single-stream Transformer visual tracking network to obtain the UAV target bounding box and tracking confidence, and converting the center position of the target bounding box into a UAV position observation in the world coordinate system, includes: Step S201: Divide the target template in the previous frame or initialization frame and the search area of ​​the current frame into image blocks using OSTrack. Perform joint feature interaction through image block embedding, position encoding and single-stream Transformer encoder to output the current target box and tracking confidence. Step S202: Based on the camera intrinsic parameter matrix and target depth, back-project the center of the target box onto the camera coordinate system, and then use the rotation matrix and translation vector from the camera to the world coordinate system to obtain the position observation in the world coordinate system.

4. The air-to-ground cooperative UAV laser wireless power supply method according to claim 1, characterized in that, Step S3, establishing a constant acceleration motion state model including the UAV's position, velocity, and acceleration, and calculating the prior state and prior covariance at the current moment based on the posterior state of the previous moment and the process noise covariance, includes: The state vector of the constant acceleration motion state model is represented as follows: ； In the formula, The three-dimensional position of the drone in motion. The three-dimensional velocity of the drone in motion. The three-dimensional acceleration of the drone in motion; The state transition matrix of the constant acceleration motion state model is represented as follows: ； In the formula, The sampling interval is... It is a 3×3 identity matrix; The mathematical expression for the prior state is: ； In the formula, This is the posterior state of the previous moment. This represents the prior state at the current moment; The mathematical expression for prior covariance is: ； In the formula, The posterior covariance of the previous time step. Let be the process noise covariance.

5. The air-to-ground cooperative UAV laser wireless power supply method according to claim 4, characterized in that, In step S4, the observation noise covariance is adaptively determined based on the tracking confidence level. and innovation saturation include: ； ； In the formula, The nominal observation noise covariance, To track confidence levels, To prevent positive numbers from being divided by zero, and These represent the minimum and maximum innovation saturation boundaries, respectively. In step S4, the a posteriori state is updated after iterative saturated Kalman filtering, and the resulting posterior motion state of the UAV includes: ； ； ； ； In the formula, For Kalman filter gain, H represents the iterative innovation quantity, and H is the observation matrix from which the position components are extracted from the state vector. For the observation of the UAV's position, For component-wise saturation operator, As the current posterior state.

6. The air-to-ground cooperative UAV laser wireless power supply method according to claim 1, characterized in that, Step S6 involves identifying the current flight phase, transition phase, or hovering phase of the UAV based on the delay compensation speed, delay compensation position, tracking confidence, and the distance between the UAV and the next preset hovering point, and calculating the ground laser access point priority coefficient and high-altitude platform priority coefficient corresponding to the current phase, including: Calculate the spatial distance between the time delay compensation position and the next preset hovering point. And calculate the magnitude of the delay compensation speed. ; when Not greater than the preset hovering distance threshold and Not greater than the preset hovering speed threshold When this occurs, it is identified as the hovering phase; when Greater than But not greater than the preset transition distance threshold ,and Greater than But not greater than the preset transition speed threshold At that time, it is identified as a transitional phase; The remaining situations are identified as the flight phase; in, ; Priority coefficient of ground laser access point during flight phase Greater than the priority coefficient of the high-altitude platform ; Priority coefficient of high-altitude platforms during hovering phase Greater than the priority coefficient of ground laser access point ; The dual-source priority factor during the transition phase is based on the progress of the drone approaching hover. The priority coefficients for the flight phase and the hovering phase are obtained by continuous interpolation.

7. The air-to-ground cooperative UAV laser wireless power supply method according to claim 6, characterized in that, Step S7, based on the priority coefficient of the ground laser access point, the priority coefficient of the high-altitude platform, the link quality of the two laser power supply links, and the UAV battery energy information, calculates the initial laser emission power of the ground laser access point and the high-altitude platform respectively, including: The initial laser emission power expression is: ； ； ； In the formula, k is h or g. This represents the initial laser emission power of the high-altitude platform. The initial laser emission power of the ground laser access point. The initial power coefficient, The maximum power of the k-th transmitter. Let be the expected received power of the k-th link. Let K be the optoelectronic link efficiency of the k-th link. For the link quality of the k-th link, It is a constant. The normalized power allocation share, To normalize charging demand, The maximum total required received power is set for the drone.

8. The air-to-ground cooperative UAV laser wireless power supply method according to claim 7, characterized in that, Step S8, which involves scaling the initial laser emission power proportionally based on the maximum allowable receiving power of the bottom-mounted photovoltaic receiver, the maximum allowable receiving power of the top-mounted photovoltaic receiver, and the maximum allowable total receiving power of the UAV, to obtain the final laser emission power of the ground laser access point and the high-altitude platform, and updating the UAV battery energy, includes: The final mathematical expression for laser emission power is: ； ； In the formula, k is either g or h. Let the final laser emission power of the k-th link be... To standardize the safety scaling factor, This represents the maximum permissible received power of the bottom-mounted photovoltaic receiver. The maximum allowable receiving power for a rooftop photovoltaic receiver is, This represents the maximum permissible total received power for the drone. After photovoltaic conversion and charging circuitry, the drone's battery energy is updated according to the following formula: ； In the formula, For photoelectric conversion efficiency, For charging efficiency, This represents the total power consumption of the drone at its current stage. To control the cycle, This represents the current battery power of the drone.

9. The air-to-ground cooperative UAV laser wireless power supply method according to claim 1, characterized in that, The process following step S4 also includes: When the tracking confidence is lower than a preset low confidence threshold, skip the observation correction and output the prior state; when the number of consecutive frames in which the tracking confidence is lower than the preset low confidence threshold reaches a preset re-detection threshold, expand the search area or reinitialize the target template.

10. A laser wireless power supply system for air-to-ground cooperative unmanned aerial vehicles, characterized in that, The system is implemented using the air-to-ground cooperative UAV laser wireless power supply method according to any one of claims 1-9, and the system includes: The status acquisition module is suitable for acquiring the spatial location information of the ground laser access point, the high-altitude platform, the UAV and the preset hovering point, and for acquiring the attitude information, battery energy information and operating status information of the UAV. The geometric modeling and link evaluation module is suitable for building three-dimensional geometric models of ground laser access points to UAVs and high-altitude platforms to UAVs, and calculating the link quality of the two laser power supply links respectively. The visual tracking and coordinate transformation module is suitable for acquiring UAV images through a high-speed camera set at the ground laser access point, inputting the current UAV image and target template into a single-stream Transformer visual tracking network to obtain the UAV target bounding box and tracking confidence, and converting the center position of the target bounding box into the UAV position observation in the world coordinate system. The confidence-aware state estimation module is suitable for establishing a constant acceleration motion state model that includes the UAV's position, speed, and acceleration. It calculates the prior state and prior covariance at the current moment based on the posterior state at the previous moment and the process noise covariance. It adaptively determines the observation noise covariance and innovation saturation bound based on the tracking confidence, and updates the posterior state after iterative saturated Kalman filtering to obtain the UAV's posterior motion state. The delay compensation and beam pointing module is suitable for obtaining the cumulative system delay based on image acquisition delay, state estimation delay, command transmission delay and turntable execution delay, propagating the UAV's posterior motion state forward along the constant acceleration motion state model to the moment after the cumulative system delay, obtaining the delay compensation position and delay compensation velocity, and calculating the azimuth and pitch commands of the ground laser access point based on the delay compensation position; The operation phase identification module is suitable for identifying whether the UAV is currently in the flight phase, transition phase, or hovering phase based on the time delay compensation speed, time delay compensation position, tracking confidence, and the distance between the UAV and the next preset hovering point, and calculating the ground laser access point priority coefficient and high-altitude platform priority coefficient corresponding to the current phase. The dual-source power allocation module is suitable for calculating the initial laser emission power of the ground laser access point and the high-altitude platform based on the priority coefficient of the ground laser access point, the priority coefficient of the high-altitude platform, the link quality of the two laser power supply links, and the battery energy information of the UAV at the current stage. The safety scaling and battery update module is suitable for scaling the initial laser emission power proportionally based on the maximum allowable receiving power of the bottom photovoltaic receiver, the maximum allowable receiving power of the top photovoltaic receiver, and the maximum allowable total receiving power of the UAV, so as to obtain the final laser emission power of the ground laser access point and the high-altitude platform and update the UAV battery energy.

Images