An unmanned aerial vehicle low-altitude interference decision management method, system, device and medium

Abstract

This invention discloses a method, system, device, and medium for low-altitude interference decision-making and management of unmanned aerial vehicles (UAVs), relating to the field of UAV technology. The method includes generating an interference perception view using real-time measured sensing data; receiving flight guidance information broadcast from a low-altitude traffic management terminal; generating candidate avoidance strategies based on parameter data extracted from the interference perception view and flight guidance information; inputting the candidate avoidance strategies into a trained prediction model for calculation to obtain the expected signal quality; performing distance matching between the location of the UAV after the execution of each candidate avoidance strategy and the conflict airspace marked in the flight guidance information, and calculating the estimated noise floor rise generated after the execution of each candidate avoidance strategy; and performing weighted analysis based on the expected signal quality, matching result, and estimated noise floor rise corresponding to each candidate avoidance strategy to determine and execute the optimal avoidance strategy, thereby improving UAV management efficiency.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) technology, and in particular to a method, system, device, and medium for low-altitude interference decision-making and management of UAVs. Background Technology

[0002] Against the backdrop of rapid development of the low-altitude economy, the scale of drone swarms continues to expand, especially with the increasing prevalence of multi-vendor equipment networking scenarios, highlighting communication interference issues.

[0003] Current mainstream interference management technologies mainly include two categories: centralized interference coordination and base station-assisted interference avoidance. Centralized technology collects interference information and makes unified decisions through a ground control center, but it has high response latency and poor scalability. Base station-assisted technology relies on the measurement capabilities of ground base stations, but it cannot function effectively at the coverage edge or in blind areas (such as tall buildings or remote airspace). Both technologies are difficult to adapt to the dynamic and complex low-altitude environment with large-scale, multi-vendor equipment.

[0004] Therefore, how to achieve effective drone decision-making and management and improve the efficiency of drone swarm management has become a technical problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0005] This invention provides a method, system, device, and medium for low-altitude interference decision management of unmanned aerial vehicles (UAVs), aiming to solve the problem of how to improve the real-time interference avoidance capability of UAV swarms in complex low-altitude environments through autonomous perception and game-theoretic decision-making by UAV terminals.

[0006] To address the aforementioned technical problems, embodiments of the present invention provide a method for low-altitude interference decision-making and management of unmanned aerial vehicles (UAVs), comprising: An interference perception view is generated using real-time measured sensing data; Receive flight guidance information broadcast from the low-altitude traffic management terminal; Based on the parameter data extracted from the interference perception view and the flight guidance information, candidate avoidance strategies are generated; The candidate avoidance strategies are input into the trained prediction model for calculation to obtain the expected signal quality achieved after executing each candidate avoidance strategy. The distance between the location of the UAV after each candidate avoidance strategy is executed and the conflict airspace marked in the flight guidance information is matched, and the estimated noise floor rise generated after each candidate avoidance strategy is calculated. The optimal avoidance strategy is determined and executed by performing a weighted analysis based on the expected signal quality, matching result, and noise floor rise prediction for each candidate avoidance strategy.

[0007] Furthermore, the process of generating the interference-perceived view includes: The system collects downlink signal data and uplink frequency band noise power data of the currently accessed base station in real time, and performs signal scanning on neighboring base stations within a preset range to obtain neighboring signal data. The flight status data of the UAV is acquired in real time, and the flight status data, downlink signal data, noise power data and neighboring cell signal data are integrated to generate the interference perception view.

[0008] Furthermore, the step of generating candidate avoidance strategies based on parameter data extracted from the interference perception view and the flight guidance information includes: Frequency band information is extracted from the interference perception view, and the frequency band information is corrected using the flight guidance information to obtain a candidate frequency band switching strategy; Based on the current UAV's transmit power and the noise power data, a candidate power adjustment strategy is generated; Based on the position data in the flight status data, an initial path offset strategy is generated, and the initial path offset strategy that satisfies the flight guidance information is used as a candidate path strategy.

[0009] Furthermore, the training process of the prediction model includes: A machine learning model is trained using historical flight records. During the training process, a loss function is constructed with the goal of minimizing the weighted error between the predicted and measured values. The prediction model is obtained by iteratively optimizing the internal parameters of the machine learning model using the loss function.

[0010] This invention provides a method for low-altitude interference decision-making and management of unmanned aerial vehicles (UAVs), comprising: Acquire approved flight plan data within the target airspace and synchronize the interference perception view reported by the UAV terminal; Based on the interference perception view and the flight plan data, the UAV's operating trajectory within the target time period is analyzed; Based on the described flight trajectory, the target airspace is gridded into a flight density heatmap; The grids that meet the preset screening conditions in the flight density heatmap are marked as conflict airspaces, and corresponding suggested communication frequency bands are assigned to different altitude layers in the flight density heatmap; The flight density heatmap, which marks conflict airspace and overlays suggested communication frequency bands, is encapsulated into flight guidance information and broadcast to each UAV terminal.

[0011] Furthermore, the step of gridding the target airspace into a flight density heatmap based on the flight trajectory includes: The target airspace is divided into grids according to a preset height level, and the running trajectory is mapped into the grid. The number of drones entering each grid within the target time period is counted. Based on the number of drones, each grid is marked according to a preset density level gradient to obtain the flight density heat map.

[0012] Furthermore, the step of marking grids that meet preset screening conditions in the flight density heatmap as conflict airspace, and allocating corresponding suggested communication frequency bands to airspaces at different altitudes in the flight density heatmap, includes: Grids in the flight density heatmap where the density level gradient exceeds a preset density threshold are marked as conflict airspace, and grids in the sensitive area of ​​the flight density heatmap are also marked as conflict airspace. Extract frequency band information from the interference perception view, and allocate suggested communication frequency bands to the airspace at different altitudes based on the frequency band information.

[0013] Another embodiment of the present invention provides a UAV low-altitude interference decision management system, including: a UAV terminal and a UAV low-altitude traffic management terminal.

[0014] Another embodiment of the present invention provides a computer device including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor executes the computer program to implement the UAV low-altitude interference decision management method as described above.

[0015] In another embodiment of the present invention, a computer-readable storage medium is provided, the computer-readable storage medium storing a computer program, wherein when the device containing the computer-readable storage medium executes the computer program, the unmanned aerial vehicle (UAV) low-altitude interference decision-making and management method described above is implemented.

[0016] Compared with the prior art, the beneficial effects of the embodiments of the present invention are at least one of the following: This invention enables local real-time interference monitoring by autonomously generating interference perception views for unmanned aerial vehicles (UAVs), reducing reliance on a central platform and improving the real-time performance of interference avoidance. Guided by received flight guidance information, the terminal acquires global airspace situational awareness, supporting collaborative decision-making. Next, candidate avoidance strategies are generated based on the view and guidance, providing diverse response options. Then, the expected signal quality is calculated using a predictive model, improving decision accuracy. Furthermore, conflict airspace matching and noise floor enhancement are combined to ensure the generated strategies comply with airspace rules. Finally, weighted analysis determines and executes the optimal strategy, enabling autonomous interference avoidance for multi-vendor UAV clusters, significantly improving the efficiency and reliability of UAV management. Attached Figure Description

[0017] Figure 1This is a schematic diagram of a drone decision management method applied to a drone terminal in one embodiment of the present invention; Figure 2 This is a schematic diagram of a drone decision management method applied to a UTM terminal in one embodiment of the present invention; Figure 3 This is a schematic diagram of the structure of a UAV low-altitude interference decision management system in one embodiment of the present invention; Figure 4 This is a structural block diagram of a preferred embodiment of a computer device provided by the present invention. Detailed Implementation

[0018] 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, and not all embodiments. The purpose of providing these embodiments is to make the disclosure of the present invention more thorough and comprehensive. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0019] In the description of this application, the terms "first," "second," "third," etc., are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Therefore, a feature defined with "first," "second," "third," etc., may explicitly or implicitly include one or more of that feature. In the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0020] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to fixed connections, detachable connections, or integral connections; they can refer to mechanical connections or electrical connections; they can refer to direct connections or indirect connections through an intermediate medium; and they can refer to the internal communication between two components. The terms "vertical," "horizontal," "left," "right," "upper," "lower," and similar expressions used herein are for illustrative purposes only and do not indicate or imply that the device or component referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as limiting the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0021] In the description of this application, it should be noted that, unless otherwise defined, all technical and scientific terms used in this invention have the same meaning as commonly understood by one of ordinary skill in the art. The terminology used in this specification is for the purpose of describing specific embodiments only and is not intended to limit the invention. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0022] One embodiment of the present invention provides a method for low-altitude interference decision-making and management of unmanned aerial vehicles (UAVs), which can be applied to UAV terminals. For details, please refer to [link to relevant documentation]. Figure 1 , Figure 1 The diagram shown is a flowchart of a drone decision management method applied to a drone terminal in one embodiment of the present invention, including the following steps: S11~S12 uses real-time measured sensing data to generate an interference sensing view and receives flight guidance information broadcast from the low-altitude traffic management terminal.

[0023] To provide accurate interference data support for subsequent game-theoretic decisions, this embodiment incorporates a sensing unit within the UAV terminal (terminal) to perform multi-dimensional signal measurements using the 3GPP TS 36.331 protocol and a fixed period. Specifically, the terminal collects downlink signal data from the currently accessed base station every 100ms in real time, including: Reference Received Power (RSRP) and Signal-to-Interference-plus-Noise Ratio (SINR). Understandably, by continuously monitoring the changes in these two indicators, the degree of interference to the current base station's downlink signal can be determined: if the RSRP value continuously decreases and the SINR value fluctuates downwards, it indicates that downlink interference may be increasing. Furthermore, uplink interference pre-assessment is conducted by measuring the noise power data of the uplink frequency band in real time, and then, combined with the terminal's own current transmit power, calculating the estimated amount of noise floor rise that may occur during uplink transmission using a local algorithm. For example, in one embodiment of the present invention, the specific calculation process is as follows: the noise floor rise prediction ΔP noise floor (dB) = 10lg(10^(Pn / 10) + 10^((Pt+Gt-Lp) / 10)) - 10lg(10^(Pn / 10)), where Pn is the uplink frequency band noise power measured in real time (unit: dBm), Pt is the current transmit power of the UAV terminal (unit: dBm), Gt is the terminal transmit antenna gain (an inherent parameter of the terminal hardware, pre-stored in the local algorithm), and Lp is the uplink signal transmission path loss (calculated based on the UAV's real-time flight altitude, distance from the access base station, and free space propagation model).

[0024] To obtain neighboring cell signal data and clarify the interference faced by each drone in its territorial area, it is also necessary to scan the signals of neighboring base stations within a preset range. For example, the terminal scans the signals of neighboring base stations within a 500-meter radius at a frequency of 1 second, only acquiring publicly available information (such as cell ID and frequency band information) that conforms to the 3GPP H1 / H2 altitude threshold measurement report standard. It is understood that this embodiment, by identifying the signal strength and distribution of these neighboring base stations, can determine potential sources of neighboring cell interference, providing data for subsequent avoidance of such interference.

[0025] After sensing the three basic signal types mentioned above, the terminal integrates them with the real-time acquired UAV flight status data to form a structured "local interference perception view." The flight status data originates from the GNSS system onboard the terminal and includes real-time position (latitude and longitude), flight altitude, and flight speed. During integration, the terminal first performs format standardization and validity checks on all types of data, eliminating invalid data with abnormal fluctuations (such as instantaneously changing RSRP values ​​or noise power values ​​exceeding reasonable ranges). Then, it binds the three signal types with the flight status data; for example, it associates the RSRP and SINR values ​​measured by the UAV at specific latitude, longitude, and altitude with the corresponding neighboring cell signal data.

[0026] The final generated local interference perception view includes: "Current Access Cell" (cell ID based on 3GPP standards, such as "PCI-123, EARFCN-37500"), "Signal Quality Matrix" (an array containing specific values ​​of RSRP and SINR, such as [-85,12]), "Uplink Interference Prediction" (an enumeration of interference levels based on noise floor measurements, such as LOW (≤3dB), MEDIUM (3-8dB), HIGH (>8dB)), "Neighboring Cell Interference List" (an array containing neighboring cell IDs and signal strengths, such as [["PCI-456", -92dBm], ["PCI-789", -105dBm]]), and "Flight Status" (an array containing position, altitude, and speed, such as [116.30°, 39.90°, 120m, 8m / s]). The "current access cell" information comes from the cell identifier obtained by the terminal when it accesses the base station, and is continuously updated during flight according to the measurement cycle specified by the 3GPP protocol.

[0027] Understandably, the drone terminal interacts with the Low-Altitude Traffic Management (UTM) system. The drone swarm receives flight guidance information broadcast by the UTM and makes strategic decisions on flight strategies under the guidance and constraints of this information to avoid interference. The flight guidance information is a standardized encapsulation of a flight density heatmap constructed by the UTM. This heatmap allows the terminal to clearly understand the overall airspace flight situation and resource (such as frequency bands and offset directions) usage recommendations.

[0028] S13. Generate candidate evasion strategies based on the parameter data extracted from the interference perception view and flight guidance information.

[0029] In this embodiment, the candidate avoidance strategies are divided into three types: frequency band switching, power adjustment, and path offset. During the candidate decision-making process, an optimal avoidance strategy needs to be selected to drive the current UAV to execute.

[0030] Regarding frequency band switching types, specific frequency band information can be extracted from the interference perception view. Specifically, frequency band information from the current access cell and neighboring cell interference lists is extracted from the interference perception view. For example, the frequency band of the current access cell is 2.6GHz, and the frequency band in the neighboring cell interference list is 4.9GHz. The current flight status of the UAV is 120m altitude. The optional parameter range for each strategy is determined using this frequency band information. Then, the flight guidance information is used to correct this frequency band information to obtain candidate frequency band switching strategies. For example, if the flight guidance information suggests "for the 100-150m altitude layer, it is recommended to use the frequency band 4.9GHz first", the following frequency band switching strategy can be generated: if there is a 4.9GHz cell in the current neighboring cell list, and the flight guidance information suggests using the 4.9GHz frequency band first, then the candidate frequency band switching strategy of 'switching to the 4.9GHz frequency band' is generated.

[0031] Regarding power adjustment types, candidate power adjustment strategies are generated based on the current UAV's transmit power and noise power data. For example: The interference level from the "Uplink Interference Prediction" is extracted from the interference perception view; that is, one of LOW (≤3dB), MEDIUM (3-8dB), or HIGH (>8dB). Simultaneously, the current UAV's own transmit power is extracted, such as 23dBm. If the minimum safe power supported by the terminal hardware is 18dBm, then the adjustable power range is 5dB. Therefore, a candidate power adjustment strategy that reduces the transmit power by 5dB can be generated accordingly. Understandably, when the uplink interference prediction level is LOW, it indicates a slight increase in noise floor, and the system can maintain the current power or make only minor adjustments to maintain link quality. When the level is MEDIUM, it indicates significant interference, and the system will trigger a power reduction mechanism to linearly reduce the transmit power in preset steps, while ensuring that it does not fall below the minimum safe power of the terminal. When the level is HIGH, it indicates severe interference, and the system will implement an aggressive power reduction strategy to reduce the transmit power to the minimum safe power limit as much as possible, so as to suppress the increase in base station noise floor to the greatest extent, thereby minimizing interference while ensuring the basic communication connection of the UAV.

[0032] Regarding path offset types, an initial path offset strategy is generated based on the position data in the flight status data. Initial path offset strategies that satisfy the flight guidance information are then used as candidate path offset strategies. For example, if the current position is 116.30°E, 39.90°N, and the heading is 90° East, an initial path offset strategy is generated within a 360-degree azimuth using this position as the origin, at fixed angular intervals (e.g., 45°) and a preset offset amount (e.g., 30 meters, less than the authorized route deviation limit of 50 meters). This initial strategy might be an offset of 30 meters from 90° East to 0° North. If the flight guidance information indicates an airspace to be avoided within 500 meters to the east, strategies that do not meet this requirement are eliminated, and candidate path offset strategies such as "adjusting the flight heading from 90° (East) to 0° (North) and offsetting horizontally by 30 meters" are generated.

[0033] S14. Input the candidate avoidance strategies into the trained prediction model for calculation to obtain the expected signal quality achieved after executing each candidate avoidance strategy.

[0034] First, a prediction model needs to be trained. In this embodiment, a machine learning model is trained using historical flight records. The machine learning model can be a multilayer perceptron or a gradient boosting decision tree (XGBoost, LightGBM).

[0035] During training, a loss function is constructed with the goal of minimizing the weighted error between the predicted and measured values. The loss function is then used to iteratively optimize the internal parameters of the machine learning model to obtain the prediction model.

[0036] Understandably, historical flight records are derived from historical data stored locally on the terminal. Each record contains RSRP, current SINR, uplink interference prediction level, a list of neighboring base station signal strengths, and historical flight guidance learned from UTM broadcasts (such as the recommended frequency band at the time). It also includes the avoidance strategy selected at the time (such as switching to the recommended 4.9G frequency band or adjusting the transmit power) and the actual SINR value after the strategy was executed.

[0037] Supervised learning is used during training, with the loss function being the weighted error between the predicted SINR and the actual SINR, as shown in the following formula: in, The number of training samples. Assigning importance weights to different scenarios (e.g., logistics scenario samples). Typical inspection scenarios ), Spatial density weights (such as high-density spatial samples) low-density airspace ), For the model to the first The predicted value for each sample, For the first The actual measurement value of each sample.

[0038] The model parameters are iteratively optimized using gradient descent. For example, in a scenario where historical data shows "the drone is at an altitude of 120m, uplink interference is MEDIUM, and the UTM broadcasts '12 drones are flying in this airspace, 4.9GHz band recommended'", selecting "switch to 4.9GHz band" improves the actual SINR from 9dB to 15dB. The training process is performed automatically while the terminal is charging, and iteration stops when the error stabilizes at ≤2dB. The trained prediction model is stored on the terminal as a lightweight file and is invoked when making interference avoidance decisions to perform millisecond-level predictions of the expected effects of multiple candidate strategies.

[0039] Based on this, during decision-making, candidate avoidance strategies generated from real-time local interference views and flight guidance information are input into the trained prediction model for calculation, driving the prediction model to output the expected signal quality, which reflects the signal quality achieved by the UAV terminal after executing each candidate avoidance strategy.

[0040] S15~S16 First, the distance between the location of the UAV after the execution of each candidate avoidance strategy and the conflict airspace marked in the flight guidance information is matched, and the noise floor rise prediction generated after the execution of each candidate avoidance strategy is calculated. Then, the expected signal quality, matching result and noise floor rise prediction corresponding to each candidate avoidance strategy are weighted and analyzed to determine the optimal avoidance strategy and execute it.

[0041] This step is the decision-making process for the optimal strategy. Specifically, it introduces three indicators—expected signal quality SINR, noise floor rise prediction, and collision spatial domain penalty—for weighted analysis to construct a utility function U. The decision objective is to maximize the utility function as the optimal avoidance strategy. In this embodiment, the utility function U is expressed as: in, , , and For weight ; This is an estimate of the increase in the noise floor of ground base stations after the strategy is implemented (e.g., during frequency band switching). ); This is the penalty value for whether a strategy is close to a conflict airspace.

[0042] It should be understood that in this embodiment, the conflict airspace represents sensitive areas that the UAV must avoid during flight, such as airports, no-fly zones, and high-density areas. Based on this, the penalty value... The determination is made by judging whether the expected position after the strategy is executed falls within the conflict airspace range broadcast by UTM. If the UAV's position will change after the path offset strategy is executed, the distance between the current position of the UAV and the conflict airspace marked in the flight guidance information needs to be calculated, and distance matching is performed according to preset rules. Specifically, if the distance display falls within the conflict airspace, a penalty value is assigned. =2, if within 50 meters of the outer edge of the conflict airspace, then... =1, a penalty value is applied if the distance is more than 50 meters outside the conflict airspace. =0.

[0043] In this embodiment, Weights representing communication quality can be selected. To meet the signal reliability requirements of low-altitude services, The weight representing the impact of uplink interference can be selected. , The weight representing conflict airspace avoidance can be set as follows: .

[0044] In another embodiment of the present invention, the strategy execution cost C can also be introduced as an indicator of the utility function U. In this case, the utility function U is expressed as follows: Selectable weights =0.2. In this embodiment, the specific value assigned to the strategy execution cost C is: frequency band switching. Power adjustment Path fine-tuning In some embodiments of the present invention, the strategy execution cost C is quantified and assigned a value based on the impact of the action type on the continuity and energy consumption of the UAV mission. Among them, power adjustment only involves the modification of baseband parameters, which takes very little time and has no risk of service interruption, so it is given the lowest weight benchmark value (e.g., 1.0); frequency band switching involves radio frequency retuning and possible brief link resynchronization, which has a risk of millisecond-level interruption, so it is given a medium weight (e.g., 2.0); path fine-tuning involves attitude calculation and trajectory replanning of the flight control system, which has the highest energy consumption and the greatest interference to the mission execution trajectory, so it is given the highest weight (e.g., 3.0).

[0045] Finally, based on the utility function formula above, the strategy that maximizes the utility function output is taken as the optimal avoidance strategy. Specifically, each drone terminal executes its corresponding optimal avoidance strategy without relying on base station or equipment manufacturer confirmation throughout the process. This enables interference avoidance while ensuring uninterrupted low-altitude services (such as logistics delivery and inspection). For example, if a drone terminal decides that its optimal avoidance strategy is to switch to the 4.9 GHz band, the terminal initiates the band switching process according to the 3GPP R17 requirements for enhanced mobility of in-flight UEs: first, disconnecting the current 2.6 GHz band connection, simultaneously scanning the base station signal of the target 4.9 GHz band, selecting the 4.9 GHz cell with the strongest signal in the neighbor cell list (such as PCI-456, RSRP=-90dBm) to initiate an access request. The entire process strictly controls the switching latency, ensuring that the switching time is ≤50ms, to avoid interruption of delivery instructions from logistics drones or video transmission from inspection drones. For example, if the optimal avoidance strategy determined by a certain drone terminal is to reduce the transmit power by 3dB, the terminal will gradually adjust the transmit power through the power control interface. After each adjustment of 1dB, the current SINR value will be retested to verify the interference mitigation effect. For instance, when the power is reduced from 23dBm to 22dBm, the retested SINR increases from 9dB to 10dB. When it is further reduced to 20dBm, the SINR stabilizes at 11dB, confirming that the power adjustment has achieved the expected result.

[0046] In some embodiments of the present invention, a verification and feedback iteration mechanism is also designed.

[0047] This embodiment illustrates the execution process of the following verification and feedback iteration mechanism: After frequency band switching, the terminal continuously measures the SINR value of the new frequency band to confirm whether it reaches the output value of the prediction model. Taking the switch to the 4.9G frequency band as an example: if the actual value deviation exceeds 2dB, other 4.9G cells are rescanned; after power adjustment, the uplink interference prediction level is retested to confirm whether it has dropped from MEDIUM to LOW. If it is still MEDIUM, the power is further reduced slightly (1dB each time); after path fine-tuning, the changes in neighboring cell signal strength are monitored to confirm whether the signals of strong interference sources in neighboring cells (such as PCI-456) in the local interference view have weakened. If the interference is not alleviated, the path offset is further increased (not exceeding the upper limit of 50 meters).

[0048] The strategy is considered effective if the actual deviation after strategy execution, the estimated uplink interference level, and the interference source meet the requirements, and the flight density in the airspace broadcast by UTM does not exceed expectations; otherwise, the strategy is considered invalid and a new decision needs to be made.

[0049] When the terminal detects that the SINR suddenly drops below 5dB (which is considered severe interference and may cause interruption of logistics instructions and video inspection stuttering), it will immediately trigger an emergency mechanism: first, reduce the transmission power to the minimum safe power, such as 18dBm, to reduce interference to the ground base station; then send an "emergency location report" (including current latitude, longitude, altitude, and interference level) value UTM to request temporary airspace priority adjustment; at the same time, quickly scan other frequency band signals in the neighboring cell list (such as other cells in 2.6G and 4.9G) and select the cell with the strongest signal and the least interference for handover.

[0050] Another embodiment of the present invention provides a method for low-altitude interference decision-making and management of unmanned aerial vehicles (UAVs), executed by a low-altitude traffic management terminal (UTM) that conforms to the 3GPP SA6 requirements for UAV identification and tracking architecture. For details, please refer to [link to relevant documentation]. Figure 2 , Figure 2 The diagram shown is a flowchart of a drone decision management method applied to a UTM terminal in one embodiment of the present invention, including: S21~S22: Obtain approved flight plan data within the target airspace, synchronize the interference perception view reported by the UAV terminal, and analyze the UAV's operating trajectory within the target time period based on the interference perception view and flight plan data.

[0051] At the Low Altitude Traffic Management (UTM) terminal, the UTM system first aggregates all approved UAV flight plan data (flight routes + timestamps) within the target airspace and synchronizes the interference perception view reported by the UAVs. Then, the UTM first parses the UAV position, altitude, and speed data in the interference perception view, and predicts the UAV's trajectory for the next 5 minutes based on this data. This trajectory specifies the location (latitude and longitude), altitude, and time of each UAV in the gridded airspace within a certain period (e.g., 5 minutes). In this embodiment, a Kalman filter algorithm can be used for prediction. For example, in the Kalman filter prediction process, firstly, the UAV motion state vector is constructed. = [x, y, z, v_x,v_y, v_z]^T, representing the three-dimensional spatial position (x, y, z) and velocity components (v_x, v_y, v_z), respectively. Next, a state transition equation is established based on a uniform motion model, using the state from the previous moment to estimate the prior position at the current moment. Then, the position, altitude, and velocity data reported in real-time by the UAV are used as observations. By calculating the Kalman gain The system dynamically adjusts the weights between the "predicted values" and the "measured values" to correct errors caused by airflow disturbances or maneuvering. Finally, it outputs the corrected posterior state estimate and uses it as a basis to recursively predict the spatiotemporal trajectory points (latitude, longitude, altitude, and timestamp) within the next 5 minutes. This process effectively smooths measurement noise and improves the accuracy of short-term trajectory prediction.

[0052] S23~S24. Based on the flight trajectory, the target airspace is gridded into a flight density heat map. Grids in the flight density heat map that meet the preset screening conditions are marked as conflict airspaces. Corresponding suggested communication frequency bands are assigned to different altitude layers in the flight density heat map.

[0053] Specifically, UTM divides the target airspace into grids based on preset altitude levels. For example, the target airspace can be divided into grids with latitude and longitude of 0.01°×0.01° and altitude of 50m as one level, such as food delivery below 120m and logistics transportation between 120-300m.

[0054] The predicted flight trajectories are mapped onto a grid, and the number of drones entering each grid within the target time period is counted. Based on the number of drones, each grid is marked according to a preset density level gradient (low, medium, high), resulting in a flight density heatmap. For example, if a grid is expected to have 8 drones entering in the next 5 minutes, that area will be marked as "medium density" on the heatmap.

[0055] During the screening process for conflict airspace, UTM identifies two categories of high-risk airspace and marks them as conflict airspace: grids with density level gradients exceeding a preset density threshold on the flight density heatmap, and grids located in sensitive areas on the flight density heatmap. For example, one category includes grids with a flight density ≥ 10 aircraft / km²; the other category includes airspace adjacent to sensitive areas such as airports and no-fly zones. Additionally, "recommended avoidance time periods" can be marked for these conflict airspaces. For instance, if an airspace is adjacent to an airport's takeoff and landing routes and flights are expected to pass between 14:05 and 14:10, it will be clearly marked that avoidance is required during that time period to prevent interference between drones and aircraft.

[0056] Furthermore, UTM extracts frequency band information from the interference perception view and allocates recommended communication frequency bands to airspace at different altitudes based on this information. Specifically, UTM, based on 3GPP's frequency band planning for low-altitude communication and combined with the frequency band information of the "currently accessed cell" in the local interference view, matches "recommended priority frequency bands" for airspace at different altitudes. For example, the local interference view displays the 100-150m altitude at which terminals frequently operate (corresponding to logistics and inspection scenarios), and, considering the advantages of the 4.9GHz band in low-altitude coverage, explicitly recommends prioritizing the use of the 4.9GHz band.

[0057] S25. Encapsulate the flight density heatmap, which marks conflict airspace and overlays suggested communication frequency bands, into flight guidance information and broadcast it to each UAV terminal.

[0058] After the above-mentioned flight density heatmap is constructed, marked and overlaid, the UTM system will broadcast the flight density heatmap to all UAV terminals in the service area through a public broadcast channel (such as the PC5 interface of LTE-V2X or a dedicated air traffic control broadcast frequency band). The broadcast process is adapted to the information receiving capabilities of the UAV terminals to ensure that the data can be effectively utilized.

[0059] Specifically, the first step is to determine the broadcast frequency. Preferably, the UTM can be set to broadcast continuously at a frequency of 2 seconds per broadcast. This frequency matches the "neighboring cell signal scan 1 second per scan" cycle of the UAV terminal. That is, when the terminal scans for neighboring cell interference every second, it can simultaneously receive the latest heat map and avoidance information, and adjust its judgment on the overall interference situation in a timely manner.

[0060] Secondly, data format standardization is implemented. In this embodiment, the broadcast content adopts a standardized JSON format, with all fields defined based on publicly available protocols to avoid incompatibility issues that could prevent terminal parsing. For example, the "Flight Density Heatmap" clearly marks the latitude and longitude range, altitude level, and estimated number of drones within the grid, while the "Suggested Frequency Band" directly labels the frequency band value (e.g., "4.9GHz"), ensuring that drone terminals of different brands and models can successfully parse the information and integrate it with their own "Local Interference Perception View." For instance, a terminal can compare the broadcast "Suggested Frequency Band 4.9GHz" with its currently accessed 2.6GHz frequency band to preliminarily determine whether a frequency band switch is necessary, laying the foundation for subsequent game theory decisions.

[0061] Finally, the broadcast coverage is controlled. In this embodiment, the broadcast range of the UTM strictly matches the low-altitude management boundary of the area to ensure that only UAVs performing tasks within this area can receive relevant information, thus avoiding UAVs in irrelevant areas from receiving redundant data.

[0062] In some embodiments of this invention, the UAV terminal locally records complete data for each strategy decision, execution, and verification feedback, periodically performs statistical analysis, and dynamically adjusts the weight coefficients of the utility function and the strategy priority to make subsequent decisions more aligned with the actual scenario. For example, if statistics show that in a scenario with an altitude of 100-150m and a flight density of ≥10 aircraft, the success rate of "frequency band switching to 4.9G" reaches 85%, while the success rate of "power adjustment" is only 50%, the "communication quality weight α" in the utility function will be fine-tuned from 0.8 to 0.85. For scenarios where the same strategy is executed multiple times with poor results (e.g., the success rate of "altitude 300m + adjacent airport airspace + path fine-tuning" is only 30%), the UAV terminal will analyze the reasons: it may be that the conflict airspace range marked by the UTM is not accurate enough. Therefore, in the airspace of an adjacent airport, path fine-tuning will no longer be prioritized; instead, the combined strategy of "frequency band switching" will be prioritized by modifying the weights.

[0063] One embodiment of the present invention provides a low-altitude interference decision and management system for unmanned aerial vehicles (UAVs). For details, please refer to [link to relevant documentation]. Figure 3 , Figure 3The diagram illustrates the structure of a UAV low-altitude interference decision management system according to one embodiment of the present invention. The system includes a UAV terminal and a UAV low-altitude traffic management terminal. Exemplarily, the system is deployed and operates within a low-altitude surveillance and service airspace (airspace range: 5km × 5km, altitude 0-300 meters) above, for example, an urban logistics park, to achieve distributed autonomous interference avoidance for a large-scale, multi-vendor UAV swarm. The UAV terminal consists of multiple UAVs (e.g., UAV1, UAV2, ... UAVn) deployed within this airspace to perform tasks such as logistics delivery and park inspection. These UAVs may be from different manufacturers, but all are equipped with hardware and software modules that meet the requirements of this system. The low-altitude traffic management terminal (UTM) can serve as a central server cluster deployed in the cloud for managing UAV low-altitude operations.

[0064] like Figure 4 As shown, this embodiment of the invention also provides a computer device. Figure 4 This is a structural block diagram of a preferred embodiment of a computer device provided by the present invention. The computer device includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the method described above.

[0065] Preferably, the computer program can be divided into one or more modules / units (such as computer program 1, computer program 2, ...), and the one or more modules / units are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units can be a series of computer program instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of the computer program in the computer device.

[0066] The processor can be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor, or the processor can be any conventional processor. The processor is the control center of the terminal device, connecting various parts of the terminal device through various interfaces and lines.

[0067] The memory mainly includes a program storage area and a data storage area. The program storage area can store the operating system, applications required for at least one function, etc., while the data storage area can store related data, etc. Furthermore, the memory can be a high-speed random access memory, or a non-volatile memory, such as a plug-in hard drive, a SmartMedia Card (SMC), a Secure Digital (SD) card, and a Flash Card, or other volatile solid-state storage devices.

[0068] It should be noted that the aforementioned terminal devices may include, but are not limited to, processors and memory, as will be understood by those skilled in the art. Figure 4 The structural block diagram is merely an example of a terminal device and does not constitute a limitation on the terminal device. It may include more or fewer components than shown, or combine certain components, or use different components. Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. The storage medium may be a magnetic disk, optical disk, read-only memory (ROM), or random access memory (RAM), etc.

[0069] Accordingly, embodiments of the present invention provide a computer-readable storage medium, the computer-readable storage medium including a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to perform the steps in the method of the above embodiments, for example... Figure 1 Steps S1 to S6 as described above.

[0070] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. A method for low-altitude interference decision-making and management of unmanned aerial vehicles (UAVs), applied to a UAV terminal, characterized in that, include: An interference perception view is generated using real-time measured sensing data; Receive flight guidance information broadcast from the low-altitude traffic management terminal; Based on the parameter data extracted from the interference perception view and the flight guidance information, candidate avoidance strategies are generated; The candidate avoidance strategies are input into the trained prediction model for calculation to obtain the expected signal quality achieved after executing each candidate avoidance strategy. The distance between the location of the UAV after each candidate avoidance strategy is executed and the conflict airspace marked in the flight guidance information is matched, and the estimated noise floor rise generated after each candidate avoidance strategy is calculated. The optimal avoidance strategy is determined and executed by performing a weighted analysis based on the expected signal quality, matching result, and noise floor rise prediction for each candidate avoidance strategy.

2. The UAV low-altitude interference decision-making and management method as described in claim 1, characterized in that, The process of generating the interference-aware view includes: The system collects downlink signal data and uplink frequency band noise power data of the currently accessed base station in real time, and performs signal scanning on neighboring base stations within a preset range to obtain neighboring signal data. The flight status data of the UAV is acquired in real time, and the flight status data, downlink signal data, noise power data and neighboring cell signal data are integrated to generate the interference perception view.

3. The UAV low-altitude interference decision-making and management method as described in claim 2, characterized in that, The step of generating candidate avoidance strategies based on parameter data extracted from the interference perception view and the flight guidance information includes: Frequency band information is extracted from the interference perception view, and the frequency band information is corrected using the flight guidance information to obtain a candidate frequency band switching strategy; Based on the current UAV's transmit power and the noise power data, a candidate power adjustment strategy is generated; Based on the position data in the flight status data, an initial path offset strategy is generated, and the initial path offset strategy that satisfies the flight guidance information is used as a candidate path strategy.

4. The UAV low-altitude interference decision-making and management method as described in claim 1, characterized in that, The training process of the prediction model includes: A machine learning model is trained using historical flight records. During the training process, a loss function is constructed with the goal of minimizing the weighted error between the predicted and measured values. The prediction model is obtained by iteratively optimizing the internal parameters of the machine learning model using the loss function.

5. A method for decision-making and management of low-altitude interference from unmanned aerial vehicles (UAVs), applied to low-altitude traffic management, characterized in that: include: Acquire approved flight plan data within the target airspace and synchronize the interference perception view reported by the UAV terminal; Based on the interference perception view and the flight plan data, the UAV's operating trajectory within the target time period is analyzed; Based on the described flight trajectory, the target airspace is gridded into a flight density heatmap; The grids that meet the preset screening conditions in the flight density heatmap are marked as conflict airspaces, and corresponding suggested communication frequency bands are assigned to different altitude layers in the flight density heatmap; The flight density heatmap, which marks conflict airspace and overlays suggested communication frequency bands, is encapsulated into flight guidance information and broadcast to each UAV terminal.

6. The UAV low-altitude interference decision-making and management method as described in claim 1, characterized in that, The step of gridding the target airspace into a flight density heatmap based on the flight trajectory includes: The target airspace is divided into grids according to a preset height level, and the running trajectory is mapped into the grid. The number of drones entering each grid within the target time period is counted. Based on the number of drones, each grid is marked according to a preset density level gradient to obtain the flight density heat map.

7. The UAV low-altitude interference decision-making and management method as described in claim 6, characterized in that, The step of marking grids that meet preset screening conditions in the flight density heatmap as conflict airspace and allocating corresponding suggested communication frequency bands to airspaces at different altitudes in the flight density heatmap includes: Grids in the flight density heatmap where the density level gradient exceeds a preset density threshold are marked as conflict airspace, and grids in the sensitive area of ​​the flight density heatmap are also marked as conflict airspace. Extract frequency band information from the interference perception view, and allocate suggested communication frequency bands to the airspace at different altitudes based on the frequency band information.

8. A decision-making and management system for low-altitude interference of unmanned aerial vehicles (UAVs), characterized in that, include: The UAV terminal that executes the UAV low-altitude interference decision management method as described in any one of claims 1 to 4 and the UAV low-altitude traffic management terminal that executes the UAV low-altitude interference decision management method as described in any one of claims 5 to 7.

9. A computer device, characterized in that, The system includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor, when executing the computer program, implements the UAV low-altitude interference decision management method as described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, wherein when the device containing the computer-readable storage medium executes the computer program, it implements the UAV low-altitude interference decision management method as described in any one of claims 1 to 7.

Images