A high-altitude long-endurance unmanned aerial vehicle connection control method and system

Abstract

The application discloses a high-altitude long-endurance unmanned aerial vehicle connection control method and system, belongs to the field of communication link management and path optimization, obtains current flight position information, attitude data and motion vector of the unmanned aerial vehicle, predicts flight trajectory change trend in a future preset time window; the trajectory trend is taken as input, a channel stability score of a candidate relay path is calculated in a ground control center in combination with a communication environment model; the current optimal path is selected according to the channel score, and a connection maintaining or switching instruction is sent to the unmanned aerial vehicle; the unmanned aerial vehicle performs connection switching after receiving the switching instruction, and feeds back switching state and connection quality parameters in real time; if the connection quality parameters are lower than a preset stability threshold, an evolutionary game algorithm is used to dynamically update the optimal path; the application improves the foresight and stability of the connection strategy, and is suitable for high-dynamic and multi-relay flight communication scenes.

Description

Technical Field

[0001] This invention relates to the field of communication link management and path optimization, specifically to a connection control method and system for high-altitude long-endurance unmanned aerial vehicles (UAVs). Background Technology

[0002] Currently, in High Altitude Long Endurance (HALE) UAV applications, such as disaster emergency communication, remote border patrol, and atmospheric data collection, missions often require UAVs to fly continuously for tens of hours or even days, maintaining a stable and reliable communication connection with the ground control center throughout this process. However, due to frequent airflow disturbances in the high-altitude environment, micro-jitter in the UAV flight path, and limited ground station / relay station resources, existing connection control mechanisms often employ fixed-frequency band channel switching or periodic reconnection strategies. These mechanisms cannot adapt to rapid fine-tuning of flight trajectories and changes in relay node load in real time, making them prone to problems such as connection delays, incorrect switching, and link drops.

[0003] Especially in complex mountainous or ocean environments, multiple relay nodes have overlapping coverage and uncertain signal blockage. Current connection control schemes cannot accurately perceive the UAV's current optimal connection direction or dynamically select the optimal relay path, resulting in low link utilization and high reconnection latency, which seriously affects mission continuity and data transmission quality.

[0004] Therefore, there is an urgent need for a connection control method suitable for high-altitude long-endurance UAVs, which can realize real-time dynamic judgment of connection optimization path and stable switching in complex flight and communication environments, so as to improve communication stability and system reliability. Summary of the Invention

[0005] The purpose of this invention is to provide a connection and control method and system for high-altitude long-endurance unmanned aerial vehicles (UAVs) to address the shortcomings in the prior art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a connection and control method for a high-altitude long-endurance unmanned aerial vehicle, comprising: The current flight position information, attitude data and motion vector of the target UAV are obtained in order to predict the flight trajectory change trend within a preset time window in the future; Using the flight trajectory change trend as input features, and combined with a preset communication environment model, the channel stability score of multiple candidate relay paths is calculated at the ground control center. Based on the channel stability score, the current optimal relay path is selected from multiple candidate relay paths, and a connection maintenance or switching command is sent to the target UAV. After receiving the switching command, the UAV performs the connection switching operation and sends the switching completion status and connection quality parameters to the ground control center in real time through the feedback link; If the connection quality parameter is lower than the preset stability threshold, the optimal connection path is dynamically updated. The channel stability score includes the channel interference coefficient calculated based on the UAV attitude change rate, historical link interruption frequency, current relay node load ratio, and the overlap between the predicted trajectory and the relay beam coverage area.

[0007] Preferably, the step of predicting the flight trajectory change trend within a future preset time window adopts a trajectory prediction model constructed based on a long short-term memory neural network or a convolutional time series network. The input of the model is a filtered sequence of flight state parameters, and the output is a set of spatial location points with timestamps.

[0008] Preferably, the channel interference coefficient in the channel stability score is calculated as follows: the cosine similarity between the attitude change rate of the UAV within the prediction time window and the main lobe direction vector of the relay beam is calculated, the perturbation function is input, and the average value of the perturbation value is obtained as the channel interference coefficient. The historical link interruption frequency is the ratio of the number of interruptions recorded by the ground control center in the relay node connection history table to the total number of connections. The current relay node load percentage is obtained by calculating the ratio of the number of currently connected drones to the maximum carrying capacity of the relay node; The overlap between the predicted trajectory and the relay beam coverage area is calculated by using the ratio of the number of trajectory points within the effective coverage area of ​​the beam cone communication in the predicted trajectory to the total number of trajectory points, which serves as the overlap factor.

[0009] Preferably, the channel stability score is a weighted sum of four factors: channel interference coefficient, historical link interruption frequency, current relay node load ratio, and trajectory overlap, with weights of 0.3, 0.25, 0.2, and 0.25, respectively, and the score value is between 0 and 1.

[0010] Preferably, the step of determining whether the current connection path is consistent with the optimal path includes: The target UAV periodically transmits the path identifier of the current connection path through the low-speed status channel, and the ground control center compares whether the identifier is consistent with the identifier corresponding to the best path in the scoring ranking.

[0011] Preferably, the steps for performing the connection switching operation include: verifying whether the received switching control command is within the set valid time window, verifying whether the hash signature of the control token matches, and completing the connection handshake, link switching and original connection disconnection process.

[0012] Preferably, the connection quality parameters include signal received strength, link bit error rate, connection delay, beam center offset angle, and initial connection stability score, and the parameters are uploaded to the ground control center through an encrypted link.

[0013] Preferably, if the connection quality parameter is lower than a preset stability threshold, the optimal connection path is updated through an evolutionary game algorithm. The evolutionary game algorithm includes initializing path payoffs, executing a replication-mutation strategy to update the path selection probability, and selecting the path with the highest probability as the target path after convergence.

[0014] Preferably, the path revenue is a weighted combination of channel stability score and relay node load ratio.

[0015] The present invention also provides a high-altitude long-endurance unmanned aerial vehicle (UAV) connection and control system, comprising: Flight status perception module: acquires the target UAV’s current flight position information, attitude data and motion vectors to predict the flight trajectory change trend within a preset time window in the future; Channel score calculation module: Taking the flight trajectory change trend as input feature, and combining it with a preset communication environment model, the channel stability score of multiple candidate relay paths is calculated at the ground control center. Optimal path decision module: Based on the channel stability score, selects the current optimal relay path from multiple candidate relay paths, and sends a connection maintenance or handover command to the target UAV; Connection switching execution module: After the UAV receives the switching command, it executes the connection switching operation and sends the switching completion status and connection quality parameters to the ground control center in real time through the feedback link; Path update control module: If the connection quality parameter is lower than the preset stable threshold, the optimal connection path is dynamically updated; The channel stability score includes the channel interference coefficient calculated based on the UAV attitude change rate, historical link interruption frequency, current relay node load ratio, and the overlap between the predicted trajectory and the relay beam coverage area.

[0016] The technical effects and advantages provided by the present invention in the above technical solution are as follows: 1. This invention provides a connection control method for high-altitude long-endurance unmanned aerial vehicles (UAVs). By constructing a dynamic communication connection control process based on flight state perception and trajectory prediction, it achieves intelligent connection decision-making and real-time switching between the UAV and relay nodes. Compared to traditional methods that rely on current location information or fixed rules for connection switching, this invention comprehensively constructs a channel stability scoring model based on multiple factors such as flight trajectory change trends, attitude disturbances, relay load, and coverage overlap. This model can identify potential unstable paths in advance, significantly improving the foresight and accuracy of the connection strategy.

[0017] 2. This invention introduces an evolutionary game theory algorithm to dynamically update the path selection strategy, overcoming the limitations of traditional static scoring mechanisms, such as slow response and high misjudgment rate, in scenarios with multiple relay resource competition. By simulating the strategy evolution process between relay paths, and combining the replication-mutation evolution mechanism and stability judgment conditions, this invention can adaptively adjust the optimal path when the connection quality deteriorates, enhancing the system's robustness to changes in complex communication environments and improving the communication continuity and system reliability of UAVs in highly dynamic flight states. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0019] Figure 1 This is a flowchart of the method of the present invention.

[0020] Figure 2 This is a flowchart of the system modules of the present invention. Detailed Implementation

[0021] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0022] Example 1, please refer to Figure 1 As shown in this embodiment, a high-altitude long-endurance unmanned aerial vehicle (UAV) connection and control method includes: The current flight position information, attitude data and motion vector of the target UAV are obtained in order to predict the flight trajectory change trend within a preset time window in the future; In a preferred embodiment of the present invention, to achieve stable communication connection control for a UAV during long-endurance high-altitude flight, it is first necessary to comprehensively perceive the flight status of the target UAV to predict its spatial position change trend over a future period, providing basic support for subsequent connection strategies. To this end, this step includes the following specific processes: The following key flight parameters at the current time are collected by the navigation sensor system deployed on the drone itself: Location information: This includes the drone's GPS coordinates (longitude, latitude, and altitude) in three-dimensional space, used to describe the drone's current location in geographic space; Attitude data, including pitch, roll, and yaw, is provided in real time by the inertial measurement unit (IMU) and is used to describe the current spatial attitude of the UAV. Motion vectors: These include velocity vectors (velocities in three axes) and acceleration vectors. This information can be obtained by fusion calculations of the IMU and GPS, reflecting the motion trend of the UAV.

[0023] The data sampling period can be set according to the system's real-time requirements, preferably between 20ms and 200ms, to ensure that the prediction model has sufficient input frequency and accuracy.

[0024] The acquired raw flight parameter data is filtered, denoised, and interpolated to reduce the impact of sensor jitter or transient errors. Preferably, an extended Kalman filter (EKF) is used to fuse multi-source heterogeneous data to enhance the stability of state estimation.

[0025] The processed flight state parameters are input into the system's preset flight trajectory prediction model for short-term trajectory extrapolation. The prediction model can be implemented in the following ways: Physical modeling methods, such as establishing state transition models based on the dynamic equations of unmanned aerial vehicles; Data-driven approaches, such as predictive models trained using Long Short-Term Memory Neural Networks (LSTM) or Convolutional Time Series Networks (TCN); Or a combination of the two, forming a hybrid trajectory prediction system.

[0026] The output of the prediction model is a sequence of predicted spatial positions of the UAV within a future preset time window (e.g., 3 to 10 seconds in the future), which is used to dynamically assess the changing trend of the spatial relative relationship between the UAV and the ground relay station.

[0027] The predicted location sequence is indexed by timestamps and outputs as a continuous set of location points P(t+Δt1), P(t+Δt2), ..., P(t+Δtn), where Δtn is the upper limit of the set trajectory prediction time window. Each location point includes spatial coordinates and a prediction error range, facilitating channel scoring and evaluation in subsequent relay path selection algorithms.

[0028] The trajectory prediction mechanism described in this step can detect the possible movement trends of the UAV in advance, overcoming the lag and one-sidedness of traditional methods that make connection decisions based on the current fixed position, and significantly improving the foresight and accuracy of connection control.

[0029] Using the flight trajectory change trend as input features, and combined with a preset communication environment model, the channel stability score of multiple candidate relay paths is calculated at the ground control center. The channel stability score includes the channel interference coefficient calculated based on the UAV attitude change rate, historical link interruption frequency, current relay node load ratio, and the overlap between the predicted trajectory and the relay beam coverage area.

[0030] After predicting the flight trajectory change trend, the system uses this trend data as a time-series input feature and combines it with a pre-established communication environment spatial model to quantitatively score multiple relay connection paths within the target flight window. This scoring process uses channel stability as the primary evaluation indicator, outputting the channel score value for each relay path within a given time period to support subsequent connection optimization decisions.

[0031] The scoring process specifically includes the calculation and integration of the following four factors: The channel interference coefficient calculated based on the attitude change rate is as follows: Let the current prediction time window be Tp, with a duration of 10 seconds; divide this time window into N equally spaced sampling points, with the sampling period denoted as Δt; at each time point ti, collect and record: pitch angle θi, roll angle Φi, and yaw angle ψi; for each instant, calculate the change between two adjacent attitude angles: Δθi = θ(i+1) - θi; ΔΦi = Φ(i+1) - Φi; Δψi = ψ(i+1) - ψi; The three-axis attitude change rate vector is defined as: Let the main beam direction vector of the relay node be... The perturbation of the signal direction by the attitude change rate is defined as: cosine similarity. Construct a perturbation function using Ci as input and calculate the perturbation value. The expression is as follows: Where k is the attenuation coefficient, with a preferred value of 4 to 6, used to adjust the sensitivity to attitude disturbances. The channel interference coefficient is defined as the average of all disturbance values, with a value range of 0 to 1. The higher the value, the more unstable the attitude and the more severe the interference to the channel.

[0032] The method for obtaining the historical link interruption frequency factor is as follows: establish a relay node connection history table to record the relay connection status of each flight mission; let the total number of historical missions be M, and the number of link interruptions be Mf; define the interruption frequency factor as: F=Mf / M; the value range is [0,1], reflecting the long-term stability of the path. Set a configurable allowable interruption threshold Fth. When F>Fth (e.g., 0.3): set the path as a "low priority" path, and its score will be weakened.

[0033] The current relay node load balancing factor is obtained as follows: For a given relay node, the number of currently connected drones is set as Uc; the maximum carrying capacity is set as Umax; the load balancing factor is defined as: L = Uc / Umax; to reduce fluctuations, a moving average method is used for smoothing. Where w is the sliding window width, the preferred value is 10 sampling points (about 1 minute). If Lavg>0.8, it means that the relay is close to congestion, and its channel score will be reduced proportionally; when real-time priority is given, the weight of this factor can be further increased.

[0034] The trajectory overlap factor reflects the geometric overlap between the future trajectory and the relay beam space, i.e., whether the UAV remains within the effective signal coverage area. For each position point Pi in the predicted trajectory sequence: determine whether Pi falls within the effective communication volume of the current relay beam cone; if it meets the cone's azimuth angle and maximum coverage radius limits, it is recorded as a valid connection point; let the total number of predicted trajectory points be N, where the number of valid connection points is Nv, and the trajectory overlap be: O = Nv / N; if O < 0.6, it is judged as a low overlap path, affecting the score; an adaptive threshold can be set to adjust the lower limit of overlap according to the flight mission requirements (range: 0.5~0.8). After normalizing the four scoring factors mentioned above, the channel stability score S is calculated using a weighted sum, as follows: The final output score S∈[0,1] is used to determine the channel reliability and priority of each relay path at the current time and during the prediction period.

[0035] Based on the channel stability score, the current optimal relay path is selected from multiple candidate relay paths, and a connection maintenance or switching command is sent to the target UAV.

[0036] To achieve stable communication connection management for high-altitude, long-endurance UAVs in complex environments, after scoring the channel stability of candidate relay paths, it is necessary to further assess the current connection status based on the scoring results and decide whether to switch paths. This process includes the following specific steps: Within the ground control center, based on the channel stability scores of all candidate relay paths calculated in the previous stage, the following operations are performed: Each relay path is labeled as Pi, and its corresponding channel stability score is Si, where i is the path number and Si∈[0,1]. Construct an ordered path rating table and sort all paths from largest to smallest according to their rating value Si; Select the path Pmax with the highest score from the sorting results. Its corresponding score is Smax. Record this path as the current optimal relay path.

[0037] If multiple relay paths have the same score, the one with the higher trajectory overlap factor will be selected first; if the trajectory overlap is still equal, the one with the lower node load ratio will be selected first.

[0038] After selecting the current optimal relay path, the ground control center needs to verify the UAV's current actual connection path to determine whether it matches the selected optimal path. The specific steps are as follows: The drone periodically transmits its current relay path unique identifier, IDcurrent, back via a low-speed status channel. Compare the current connection path identifier IDcurrent with the obtained optimal path identifier IDmax; If IDcurrent=IDmax, it means that the current connection is the optimal path and there is no need to switch; generate a connection keep-alive instruction. If they are not equal, proceed to the switch preparation phase and execute the next step.

[0039] The unique identifier of the connection path is a 32-bit static binary code, and the code structure includes the relay node number, the geographic location block number, and the beam index number.

[0040] When the determination result indicates that the current connection path is inconsistent with the optimal path, in order to ensure the uniqueness and timeliness of the path switching operation, an authorization token with a timestamp needs to be generated and bound to the target path information. The specific process is as follows: Based on the current ground control center time Tgen, generate a timestamp with an encrypted hash signature; use a hash function (such as SHA-256) to encrypt and concatenate the following fields to form a switching authorization token Tokenswitch: timestamp Tgen, current optimal path identifier IDmax, and control command code (set as "CMD_SWITCH"); bind this token with the path information to generate a complete control command data packet, with the following data structure:

[0041] To ensure the security and timeliness of command transmission, the switching control command must be sent to the target UAV through a dedicated secure signaling channel. The specific execution method is as follows: The ground control center synchronously sends the aforementioned control command data packets to the target UAV's main control communication unit via a signaling link with an encrypted authentication mechanism; After receiving the instruction, the drone first verifies the validity of the authorization token, including: Verify that the timestamp is within a 10-second window of the current time; Check if the target path identifier in the comparison token exists in the current candidate path list; Verify whether the hash signature matches the preset key; After successful verification, the main control communication unit initiates the switching process, gradually disconnecting the current relay connection and establishing a new connection with the relay node of the target path without affecting the flight mission. After the switch is completed, the system will confirm the successful path switch to the ground control center via the feedback channel in the next status feedback cycle and report the current connection status.

[0042] By following the steps above, relay paths can be dynamically evaluated and switched while ensuring command security and connection stability, thus maintaining the UAV's communication link in optimal condition throughout flight. Compared to traditional periodic switching methods, this method offers advantages such as strong real-time response capabilities, low false switching rate, and richer path determination criteria.

[0043] After receiving the handover command, the UAV performs the connection handover operation and sends the handover completion status and connection quality parameters to the ground control center in real time via the feedback link.

[0044] After the ground control center selects the optimal relay path and issues a connection switching command to the target UAV, the target UAV needs to perform the connection switching operation based on the received control data and return the switching completion status and current connection quality parameters to the ground control center via the feedback link for subsequent communication status maintenance and score updates. The specific steps are as follows: After receiving the switching command data packet from the ground control center, the UAV main control communication unit first verifies the validity of the command. The verification includes: Whether the instruction is within the authorized valid time window (preferably within 10 seconds after receipt); Check if the hash signature of the switching authorization token matches the local preset key; Does the target relay path identifier exist in the list of currently available paths? Is the current connection state allowed to be switched (e.g., during non-critical communication periods)?

[0045] After successful verification, the connection switching process begins.

[0046] Preparation phase: Load the target path identifier into the temporary connection cache queue; initialize the handshake process with the target relay node, including frequency matching, modulation parameter synchronization and beam alignment request; set the connection timeout threshold, preferably 3 seconds, otherwise the handover is considered a failure.

[0047] Switching phase: Without affecting the flight control system and mission payload communication, gradually reduce the transmit power of the current connection; after completing the handshake confirmation with the target relay node, officially switch the main communication link to the target path; clear the channel buffer and session context information of the original connection to prevent residual interference from the old link.

[0048] Confirmation phase: Check whether the new link has completed connection registration, IP address allocation (if using IP link layer), and channel quality meets the standards; if all initialization steps are completed within the preset time, mark the handover operation as successful; if the handover fails (e.g., handshake response timeout, target relay node unavailable), trigger the rollback process, restore the original connection state, and report the abnormal event.

[0049] During the first communication cycle after the connection switch is completed, the drone needs to assess the current connection status in real time and collect the following parameters: Received Signal Strength Indicator (RSSI): Measured in dBm, it reflects the current received signal power level; Link error rate: The percentage of data packets that are received per unit of time; Connection latency: measures the round-trip delay of signaling; Relay beam coverage center offset: Calculates the angular deviation between the current UAV position and the center of the target relay beam main axis, in degrees; Initial connection stability estimate: After normalizing the above indicators, the initial connection stability score is calculated according to the preset function and used for subsequent scoring model updates.

[0050] The above parameters will be encapsulated into status data packets and uploaded to the ground control center in real time via an encrypted communication channel.

[0051] The structure of the feedback data packet is defined as follows:

[0052] The optimal feedback upload period is between 500 milliseconds and 2 seconds, which can be dynamically adjusted according to the actual task priority.

[0053] Through the aforementioned connection switching operation and feedback mechanism, real-time optimal adjustment and self-closed-loop feedback control of the UAV communication path are achieved. This method avoids the link jitter, repeated switching, and mission interruption problems caused by delayed confirmation or lack of feedback in traditional path switching, significantly improving the reliability of the UAV communication link and mission continuity under highly dynamic flight conditions.

[0054] If the connection quality parameter is lower than a preset stable threshold, the optimal connection path is dynamically updated.

[0055] When the connection quality parameters of the target UAV's current connection path are lower than a preset stability threshold, the ground control center will no longer directly select a path based on the current score. Instead, it will introduce an evolutionary game theory algorithm to simulate the strategy competition process among multiple relay paths, discovering the optimal path from dynamic evolution. This algorithm does not rely on static scores but is based on the stability of strategy evolution and dynamic payoff changes among paths, making it suitable for complex scenarios involving resource competition among multiple relay nodes and frequent fluctuations in the communication environment. Its specific implementation process includes the following steps: All available relay paths within the prediction time window will be labeled as P1, P2, ..., Pn, where n is the number of candidate paths; Each relay path is treated as an independent policy participant, and its initial policy reward value Ri is defined as a weighted combination of the current channel stability score and the relay node load ratio of that path, calculated as follows: Let the channel stability score be Si, and the relay node load ratio be Li; the revenue function is defined as: Wherein, α and β are adjustable weight coefficients, satisfying α+β=1, preferably set to α=0.7 and β=0.3. The strategy reward value Ri represents the path's ability to attract resources during evolution and is used to guide the evolution of path strategies.

[0056] Initialize the path strategy distribution vector X={x1,x2,...,xi}, where xi represents the probability that path Pi is currently selected, and all xi sum to 1. The evolutionary game process, based on replication and mutation, simulates the competitive behavior of paths in multiple communication resource allocation cycles and updates path selection probabilities, specifically including: Copying operation: Update probabilities according to the relative payoff of each path. If the payoff of path Pi is higher than the average, its probability of being selected in the next round increases. The update method is as follows: Wherein, γ is the learning rate, preferably 0.1; This represents the average profit across all paths.

[0057] Mutation operation: To prevent getting trapped in local optima, a small amount of mutation perturbation is introduced after each round of evolution, using the following method: Where δ is the variation ratio, preferably 0.05; ϵi is a randomly generated small probability value that satisfies the normalization condition.

[0058] The above process is iterated for several rounds (preferably 5 to 10 rounds) until the path selection probability distribution is stable or the maximum number of rounds is reached.

[0059] Calculate the rate of change of the path selection probability distribution after each round of evolution: If the rate of change of the selection probability of all paths is lower than the set convergence threshold (e.g., 0.01), then... If the condition holds true for all i, then the current system is determined to have reached an evolutionary stable state. Choose the path Pbest with the highest selection probability from the path strategy in the steady state as the updated optimal connection path; If a stable state has not been reached but the maximum number of evolution rounds has been exceeded, the path with the highest current reward value is directly selected as the compensation decision to ensure real-time performance.

[0060] The updated optimal connection path Pbest's path identifier is encapsulated with the switching control command to generate a data packet containing the following fields: Target path identifier; command type (connection switch); generation timestamp; encrypted authentication token. The data packet is sent to the target UAV's communication unit via a secure communication link from the ground control center. After receiving the instruction and completing the verification, the drone initiates the connection switching process and enters the feedback process, reporting the switching completion status and the connection quality parameters of the new path.

[0061] By introducing an evolutionary game theory algorithm, this method can simulate the competitive behavior among relay paths in dynamic environments where connection quality deteriorates, achieving "collective intelligence" optimization of path selection. Compared to static scoring or traditional filtering algorithms, this method has stronger adaptability and anti-interference capabilities, and is particularly suitable for highly dynamic flight scenarios with frequent changes in relay nodes or uneven loads.

[0062] Example 2, please refer to Figure 2 As shown in this embodiment, a high-altitude long-endurance unmanned aerial vehicle (UAV) connection and control system includes: Flight status perception module: acquires the target UAV’s current flight position information, attitude data and motion vectors to predict the flight trajectory change trend within a preset time window in the future; Channel score calculation module: Taking the flight trajectory change trend as input feature, and combining it with a preset communication environment model, the channel stability score of multiple candidate relay paths is calculated at the ground control center. Optimal path decision module: Based on the channel stability score, selects the current optimal relay path from multiple candidate relay paths, and sends a connection maintenance or handover command to the target UAV; Connection switching execution module: After the UAV receives the switching command, it executes the connection switching operation and sends the switching completion status and connection quality parameters to the ground control center in real time through the feedback link; Path update control module: If the connection quality parameter is lower than the preset stable threshold, the optimal connection path is dynamically updated; The channel stability score includes the channel interference coefficient calculated based on the UAV attitude change rate, historical link interruption frequency, current relay node load ratio, and the overlap between the predicted trajectory and the relay beam coverage area.

[0063] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A connection and control method for a high-altitude long-endurance unmanned aerial vehicle (UAV), characterized in that: include: The current flight position information, attitude data and motion vector of the target UAV are obtained in order to predict the flight trajectory change trend within a preset time window in the future; Using the flight trajectory change trend as input features, and combined with a preset communication environment model, the channel stability score of multiple candidate relay paths is calculated at the ground control center. Based on the channel stability score, the current optimal relay path is selected from multiple candidate relay paths, and a connection maintenance or switching command is sent to the target UAV. After receiving the switching command, the UAV performs the connection switching operation and sends the switching completion status and connection quality parameters to the ground control center in real time through the feedback link; If the connection quality parameter is lower than the preset stability threshold, the optimal connection path is dynamically updated. The channel stability score includes the channel interference coefficient calculated based on the UAV attitude change rate, historical link interruption frequency, current relay node load ratio, and the overlap between the predicted trajectory and the relay beam coverage area.

2. The connection and control method for a high-altitude long-endurance unmanned aerial vehicle according to claim 1, characterized in that: The step of predicting the flight trajectory change trend within a future preset time window adopts a trajectory prediction model constructed based on a long short-term memory neural network or a convolutional time series network. The input of the model is a filtered sequence of flight state parameters, and the output is a set of spatial location points with timestamps.

3. The connection and control method for a high-altitude long-endurance unmanned aerial vehicle according to claim 1, characterized in that: The channel interference coefficient in the channel stability score is calculated as follows: the cosine similarity between the attitude change rate of the UAV within the prediction time window and the main lobe direction vector of the relay beam is calculated, the perturbation function is input, and the average value of the perturbation value is obtained as the channel interference coefficient. The historical link interruption frequency is the ratio of the number of interruptions recorded by the ground control center in the relay node connection history table to the total number of connections. The current relay node load percentage is obtained by calculating the ratio of the number of currently connected drones to the maximum carrying capacity of the relay node; The overlap between the predicted trajectory and the relay beam coverage area is calculated by using the ratio of the number of trajectory points within the effective coverage area of ​​the beam cone communication in the predicted trajectory to the total number of trajectory points, which serves as the overlap factor.

4. The connection and control method for a high-altitude long-endurance unmanned aerial vehicle according to claim 3, characterized in that: The channel stability score is a weighted sum of four factors: channel interference coefficient, historical link interruption frequency, current relay node load ratio, and trajectory overlap, with weights of 0.3, 0.25, 0.2, and 0.25, respectively. The score value is between 0 and 1.

5. The connection and control method for a high-altitude long-endurance unmanned aerial vehicle according to claim 1, characterized in that: The steps to determine whether the current connection path is consistent with the optimal path include: The target UAV periodically transmits the path identifier of the current connection path through the low-speed status channel, and the ground control center compares whether the identifier is consistent with the identifier corresponding to the best path in the scoring ranking.

6. The connection and control method for a high-altitude long-endurance unmanned aerial vehicle according to claim 1, characterized in that: The steps for performing a connection handover operation include: verifying whether the received handover control command is within the set valid time window, verifying whether the hash signature of the control token matches, and completing the connection handshake, link handover, and original connection disconnection process.

7. The connection and control method for a high-altitude long-endurance unmanned aerial vehicle according to claim 1, characterized in that: Connection quality parameters include signal received strength, link bit error rate, connection delay, beam center offset angle, and initial connection stability score. These parameters are uploaded to the ground control center via an encrypted link.

8. The high-altitude long-endurance UAV connection and control method according to claim 7, characterized in that: If the connection quality parameter is lower than a preset stability threshold, the optimal connection path is updated through an evolutionary game algorithm. The evolutionary game algorithm includes initializing path payoffs, executing a replication-mutation strategy to update the path selection probability, and selecting the path with the highest probability as the target path after convergence.

9. A connection and control method for a high-altitude long-endurance unmanned aerial vehicle according to claim 8, characterized in that: The path revenue is a weighted combination of channel stability score and relay node load percentage.

10. A high-altitude long-endurance unmanned aerial vehicle (UAV) connection and control system, used to implement the high-altitude long-endurance UAV connection and control method according to any one of claims 1-9, characterized in that: include: Flight status perception module: acquires the target UAV’s current flight position information, attitude data and motion vectors to predict the flight trajectory change trend within a preset time window in the future; Channel score calculation module: Taking the flight trajectory change trend as input feature, and combining it with a preset communication environment model, the channel stability score of multiple candidate relay paths is calculated at the ground control center. Optimal path decision module: Based on the channel stability score, selects the current optimal relay path from multiple candidate relay paths, and sends a connection maintenance or handover command to the target UAV; Connection switching execution module: After the UAV receives the switching command, it executes the connection switching operation and sends the switching completion status and connection quality parameters to the ground control center in real time through the feedback link; Path update control module: If the connection quality parameter is lower than the preset stable threshold, the optimal connection path is dynamically updated; The channel stability score includes the channel interference coefficient calculated based on the UAV attitude change rate, historical link interruption frequency, current relay node load ratio, and the overlap between the predicted trajectory and the relay beam coverage area.

Images