An unmanned aerial vehicle autonomous return method and flight control device in an interference environment

Abstract

The application discloses a kind of unmanned aerial vehicle self-return method and flight control device under interference environment, it is related to unmanned aerial vehicle return technical field, for reducing the risk of unmanned aerial vehicle yaw loss.In the method, in the interference environment that unmanned aerial vehicle encounters communication and satellite navigation double failure, avoid the cumulative error and yaw loss risk caused by blind long-distance return.By introducing active signal search mode, priority is tried to restore communication link outside local interference area, when search is fruitless or power is limited, reverse rollback is carried out using pre-recorded feature waypoint and multi-source fusion navigation, and emergency safety airway is cut in, so that the risk of unmanned aerial vehicle yaw loss is reduced in complex electromagnetic environment.

Description

Technical Field

[0001] This application relates to the field of unmanned aerial vehicle (UAV) return-to-home technology, and in particular to a method and flight control device for autonomous UAV return-to-home in a jammed environment. Background Technology

[0002] With the widespread application of drones in military reconnaissance, power line inspection, logistics delivery, and emergency rescue, the issues of drone flight safety and autonomous control have become increasingly prominent. During routine missions, drones typically rely on satellite navigation systems (such as GPS) to obtain high-precision self-position information and on real-time wireless communication links established with ground stations to receive flight control commands and transmit telemetry data, thereby ensuring the smooth progress of flight missions.

[0003] To address potential communication link interruptions during flight, relevant unmanned aerial vehicle (UAV) systems typically employ a fixed-track autonomous return-to-home (ORW) technology. Specifically, the UAV records its takeoff coordinates and historical waypoints in real time during takeoff or flight. Upon detecting a loss of control signal from the ground station, a pre-set passive return-to-home mechanism is immediately triggered. This mechanism directly utilizes real-time positioning data from the satellite navigation module to control the UAV to climb to a preset safe altitude and fly in a straight line towards the return-to-home endpoint, or return along the recorded mission route.

[0004] However, when drones encounter localized interference environments where communication and satellite navigation are simultaneously restricted, the relevant technologies will still drive the drones to perform long-distance physical return flights, leaving the drones in a state of blind flight without communication control for a long time. This causes the trajectory calculation error to accumulate continuously with the flight distance, thereby increasing the risk of the drones veering off course and getting lost. Summary of the Invention

[0005] This application provides a method and flight control device for autonomous return of unmanned aerial vehicles (UAVs) in a jammed environment, which can reduce the risk of UAVs veer off course and become lost.

[0006] Firstly, a method for autonomous return of a UAV under interference conditions is provided, characterized by its application to flight control equipment. The method includes: acquiring a pre-generated preset emergency safety route; and, when the UAV is performing a flight mission and the satellite navigation signal and communication link are normal, real-time monitoring of the UAV's flight attitude change data and flight mode switching status; when the flight attitude change data exceeds a preset attitude change threshold or the flight mode switching status changes, recording the current position coordinates and signal strength data as feature waypoints; when a communication link interruption and satellite navigation signal loss are detected, calling a multi-source fusion navigation algorithm to output the current pose estimate, and determining the current interference level based on a preset comprehensive signal quality index; and when the current interference level is determined to be localized severe interference and the UAV... If the remaining battery power is greater than a first preset threshold, the drone is controlled to enter a signal search mode. In signal search mode, the drone maintains its positioning based on a multi-source fusion navigation algorithm and performs active signal search flight in the direction of signal enhancement based on the spatial signal strength gradient. If the communication link is not restored during active signal search flight and the preset mode switching conditions are met, the signal search mode is interrupted and the drone is controlled to switch to a safe return mode. The preset mode switching conditions include the search duration exceeding a preset duration or the remaining battery power dropping to a second preset threshold. In safe return mode, the drone is controlled to fly in reverse along the feature waypoints based on the pose estimation value to the target feature waypoint that matches the preset emergency safety route space, and then enters the preset emergency safety route from the target feature waypoint to perform the return flight.

[0007] By adopting the above technical solution, in interference environments where both communication and satellite navigation fail, the cumulative error and risk of yaw and disorientation caused by blindly returning over long distances are avoided. By introducing an active signal search mode, priority is given to attempting to restore communication links outside the local interference zone. When the search is unsuccessful or battery power is limited, pre-recorded characteristic waypoints and multi-source fusion navigation are used for reverse backtracking, and an emergency safety route is initiated. This reduces the risk of yaw and disorientation of the drone in complex electromagnetic environments.

[0008] In conjunction with some embodiments of the first aspect, in some embodiments, the step of recording the current position coordinates and signal strength data as characteristic waypoints when the flight attitude change data is greater than a preset attitude change threshold or the flight mode switching state changes, specifically includes: real-time monitoring of the signal-to-noise ratio (SNR) data of the airborne communication module; when the SNR at the current moment is greater than the historical average SNR within a preset time window, and the difference between the current SNR and the historical average SNR is greater than a preset gain threshold, marking the spatial point corresponding to the current moment as a first candidate waypoint; calculating the terrain obstruction angle between the current position and a known ground station based on the airborne pre-stored digital elevation model; marking the corresponding spatial point as a second candidate waypoint at the zero-crossing moment when the terrain obstruction angle changes from positive to negative; marking the spatial point corresponding to the current moment as a third candidate waypoint when the flight attitude change data is greater than the preset attitude change threshold or the flight mode switching state changes; and filtering the first, second, and third candidate waypoints based on a preset spatial resolution to obtain characteristic waypoints.

[0009] By adopting the above technical solution, the first candidate waypoint is extracted by real-time monitoring of the signal-to-noise ratio data of the communication module, the second candidate waypoint is extracted by calculating the zero-crossing time of the terrain occlusion angle using a digital elevation model, and the third candidate waypoint is extracted by combining flight attitude and mode changes. Finally, the candidate points are filtered and deduplicated based on preset spatial resolution and signal strength. This results in a final sequence of feature waypoints that not only includes the geometric features of the trajectory but also integrates spatial electromagnetic distribution and terrain occlusion information. This provides a more accurate and valuable path navigation foundation for UAVs to search for signals and safely retreat in interference environments.

[0010] In conjunction with some embodiments of the first aspect, in some embodiments, the step of filtering the first candidate track point, the second candidate track point, and the third candidate track point based on a preset spatial resolution to obtain a feature track point specifically includes: calculating the spatial Euclidean distance between any two candidate track points; comparing the signal strength data corresponding to the two candidate track points when the spatial Euclidean distance is less than the preset spatial resolution; retaining the candidate track point with the larger signal strength data as the feature track point, and removing the other candidate track point.

[0011] By adopting the above technical solution, spatial deduplication of dense candidate waypoints is performed using a preset spatial resolution, and waypoints with better signal quality are retained first when they are close to each other, thereby controlling the amount of data stored for feature waypoints and reducing the memory usage of flight control equipment.

[0012] In conjunction with some embodiments of the first aspect, in some embodiments, during active signal search flight, the method further includes: acquiring in real time the position error covariance matrix representing the positioning uncertainty output by the multi-source fusion navigation algorithm; calculating the spatial backoff distance based on the pose estimation value at the current moment and the position coordinates of the nearest feature track point; calculating the maximum allowable drift threshold by combining the safe capture radius of the nearest feature track point and the spatial backoff distance; and triggering a preset mode switching condition when the real-time cumulative drift error corresponding to the position error covariance matrix reaches the maximum allowable drift threshold.

[0013] By adopting the above technical solution, the real-time positioning uncertainty of the multi-source fusion navigation algorithm is incorporated into the return-to-home decision mechanism. By dynamically calculating the maximum allowable drift threshold, the system can more promptly stop fruitless signal searches and force entry into the safe return-to-home mode before the navigation error accumulates to the point where historical waypoints cannot be accurately found, thus preventing complete disorientation caused by prolonged blind flight.

[0014] In conjunction with some embodiments of the first aspect, in some embodiments, the step of calculating the maximum permissible drift threshold by combining the safe capture radius of the nearest feature waypoint with the spatial back-off distance specifically includes: obtaining the preset prior drift rate of the multi-source fusion navigation algorithm in the absence of satellite navigation; multiplying the spatial back-off distance by the preset prior drift rate to obtain the expected back-off drift increment required for the UAV to fly over the spatial back-off distance; and subtracting the expected back-off drift increment from the safe capture radius to obtain the maximum permissible drift threshold at the current moment.

[0015] By adopting the above technical solution and assessing the impact of the current position uncertainty of the UAV on the subsequent retreat mission, a more accurate maximum allowable drift threshold was obtained.

[0016] In conjunction with some embodiments of the first aspect, in some embodiments, the step of performing active signal search flight based on the spatial signal strength gradient towards the signal enhancement direction specifically includes: controlling the UAV to perform a preset detection maneuver to obtain the signal strength of multiple points in space; extracting the three-dimensional pointing vector of the current spatial signal strength gradient based on the signal strength of the multiple points in space; extracting a local two-dimensional elevation profile line along the three-dimensional pointing vector in the onboard pre-stored digital elevation model, with the current position of the UAV as the starting point and a preset detection distance as the ending point; calculating the elevation angle of each sampling point on the local two-dimensional elevation profile line relative to the current position of the UAV; if the elevation angle of any sampling point is greater than the sum of the current flight elevation angle of the UAV and the preset safety margin, determining that the three-dimensional pointing vector is physically blocked, and marking the corresponding spatial signal strength gradient as a pseudo-signal gradient caused by multipath reflection and removing it; controlling the UAV to perform active signal search flight along the signal gradient direction after removing the pseudo-signal gradient.

[0017] By adopting the above technical solution, a digital elevation model is introduced to verify terrain occlusion when conducting active search based on signal strength gradient. This enables the identification and elimination of multipath reflection pseudo-signal gradients caused by physical terrain occlusion, preventing the UAV from being guided to the wrong direction or crashing into obstacles by false signals, thus improving the safety and optimization efficiency of signal search flight.

[0018] In conjunction with some embodiments of the first aspect, in some embodiments, the preset emergency safety route includes pre-marked communication reachable points and emergency landing points. After the step of entering the preset emergency safety route from the target feature track point to perform the return flight, the method further includes: reducing the flight speed to attempt to re-establish the communication link while flying along the preset emergency safety route and passing through the communication reachable points; and if the remaining battery power is lower than a third preset threshold during the flight along the preset emergency safety route, controlling the UAV to fly to the emergency landing point closest to the current position, and using the onboard visual sensor to detect the ground flatness to perform a vertical safe landing.

[0019] By adopting the above technical solutions, the drone can actively reduce its speed when passing through a communication reachable point to increase the probability of reconnection. When the battery is severely low, it can select the nearest alternative landing point and land safely using visual sensors. This further improves the safety mechanism in extreme situations and reduces the risk of drone crashes.

[0020] In a second aspect, embodiments of this application provide a flight control device, which includes: one or more processors and a memory; the memory is coupled to the one or more processors, and the memory is used to store computer program code, which includes computer instructions, and the one or more processors call the computer instructions to cause the flight control device to perform the method described in the first aspect and any possible implementation thereof.

[0021] Thirdly, embodiments of this application provide a computer program product containing instructions that, when the computer program product is run on a flight control device, cause the flight control device to execute the method described in the first aspect and any possible implementation thereof.

[0022] Fourthly, embodiments of this application provide a computer-readable storage medium including instructions that, when executed on a flight control device, cause the flight control device to perform the method described in the first aspect and any possible implementation thereof.

[0023] Understandably, the flight control equipment provided in the second aspect, the computer program product provided in the third aspect, and the computer storage medium provided in the fourth aspect are all used to execute the methods provided in the embodiments of this application. Therefore, the beneficial effects they can achieve can be referred to the beneficial effects in the corresponding methods, and will not be repeated here.

[0024] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages:

[0025] 1. In environments where both communication and satellite navigation fail, drones can avoid the cumulative errors and the risk of yaw and disorientation caused by blindly returning over long distances. By introducing an active signal search mode, priority is given to attempting to restore communication links outside the local interference zone. When the search fails or battery power is limited, pre-recorded characteristic waypoints and multi-source fusion navigation are used for reverse backtracking, and an emergency safety route is initiated. This reduces the risk of drone yaw and disorientation in complex electromagnetic environments.

[0026] 2. By real-time monitoring of the signal-to-noise ratio data of the communication module, the first candidate waypoint is extracted. The second candidate waypoint is extracted by calculating the zero-crossing time of the terrain obstruction angle using a digital elevation model. The third candidate waypoint is extracted by combining flight attitude and mode changes. Finally, the candidate points are filtered and deduplicated based on preset spatial resolution and signal strength. This results in a final sequence of feature waypoints that not only includes the geometric features of the trajectory but also integrates spatial electromagnetic distribution and terrain obstruction information. This provides a more accurate and valuable path navigation foundation for UAVs to search for signals and safely retreat in interference environments.

[0027] 3. When conducting active search based on signal strength gradient, a digital elevation model is introduced to verify terrain occlusion. This enables the identification and elimination of multipath reflection pseudo-signal gradients caused by physical terrain occlusion, preventing the UAV from being guided in the wrong direction or crashing into obstacles by false signals, thus improving the safety and optimization efficiency of signal search flight. Attached Figure Description

[0028] Figure 1 This is a flowchart illustrating a method for autonomous return of a drone under interference conditions, as described in an embodiment of this application.

[0029] Figure 2 This is another flowchart illustrating a method for autonomous return of a drone under interference conditions, as described in this application.

[0030] Figure 3 This is a schematic diagram of the physical device structure of a flight control device in the embodiments of this application. Detailed Implementation

[0031] The terminology used in the following embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. As used in the specification and appended claims of this application, the singular expressions “a,” “an,” “the,” “the,” “the,” and “this” are intended to include the plural expressions as well, unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in this application refers to and includes any or all possible combinations of one or more of the listed items.

[0032] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature, and in the description of the embodiments of this application, unless otherwise stated, "multiple" means two or more.

[0033] This application provides a method and flight control device for autonomous return of unmanned aerial vehicles (UAVs) in a jammed environment, which can reduce the risk of UAVs veer off course and become lost.

[0034] Please see Figure 1 This is a flowchart illustrating a method for autonomous return of a drone under interference conditions in an embodiment of this application.

[0035] S101. Obtain the pre-generated preset emergency safety route, and monitor the UAV's flight attitude change data and flight mode switching status in real time when the UAV is performing a flight mission and the satellite navigation signal and communication link are normal.

[0036] Among them, the preset emergency safety route refers to one or more safe routes from the mission area back to a preset destination (such as the takeoff point or alternate landing point) that are pre-calculated and stored in the UAV's non-volatile memory before the UAV mission is executed, based on multi-dimensional information such as digital elevation models, known obstacles, historical signal coverage maps, and no-fly zones, using path planning algorithms. Flight attitude change data is used to represent the rotational motion state of the UAV during flight, typically including changes in angle, angular velocity, or angular acceleration around its three principal axes (roll, pitch, and yaw), output by sensors such as inertial measurement units (IMUs). Flight mode switching refers to the transition of the UAV flight control equipment between different control logics, such as switching from autonomous cruise mode to hover mode, or from manual control mode to mission route mode.

[0037] Specifically, the flight control equipment first loads preset emergency safety flight path data from its internal storage (such as Flash or SD card) into memory for quick retrieval when needed. Simultaneously, the flight control equipment initiates a high-frequency monitoring task, which continuously reads real-time data from onboard sensors (especially the IMU) and the flight control state machine. Monitoring changes in flight attitude aims to capture critical moments when the UAV performs maneuvers such as turns, climbs, and descents. Monitoring flight mode switching aims to identify changes in flight behavior patterns triggered by ground station commands or onboard autonomous logic.

[0038] S102. When the flight attitude change data exceeds the preset attitude change threshold or the flight mode switching state changes, record the current position coordinates and signal strength data as feature track points.

[0039] Among them, the preset attitude change threshold refers to the threshold for the amount or rate of change of attitude angles pre-set to determine whether the UAV has performed significant maneuvers, such as a yaw angle change exceeding 30 degrees or a pitch angular velocity greater than 15 degrees / second. Position coordinates refer to the spatial position of the UAV at a given moment in a specific coordinate system (usually the WGS-84 geographic coordinate system), consisting of longitude, latitude, and altitude components. Signal strength data represents the quality parameters of the wireless communication signals received by the UAV from the ground station or relay, typically Received Signal Strength Indication (RSSI) or Signal-to-Noise Ratio (SNR). Featured waypoints are key spatial points in the UAV's flight path selected according to specific rules that characterize the geometry and communication environment of the flight path; they are the basic units for constructing the original return path.

[0040] Specifically, when the cumulative amount of attitude changes (such as turning or pitching) of the drone exceeds a preset attitude change threshold, or when its flight mode changes, the current moment is determined to be a critical moment worth recording. These moments often correspond to changes in heading, transitions in mission phases, or responses to specific environments; they are geometric inflection points or logical nodes in the flight path. Recording waypoints at these moments, compared to recording at fixed time intervals, allows for a more accurate and reliable reproduction of a safe and effective flight path with less data. Once the cumulative amount of attitude changes (such as turning or pitching) of the drone exceeds the preset attitude change threshold, or when its flight mode changes, the flight control equipment immediately obtains the current position coordinates from the navigation system (in this case, GPS) and the current signal strength data from the communication module. This information, along with auxiliary data such as timestamps, is packaged into a data structure and stored in a queue or list specifically for storing characteristic waypoints.

[0041] In some embodiments, the recording of feature waypoints in this step can be achieved in several ways. Optionally, a trigger logic based on multiple conditions can be used to record feature waypoints. First, multiple independent trigger conditions are set, such as a single change in yaw angle greater than 30 degrees, a cumulative change in altitude greater than 20 meters, and a flight mode switching from cruise to hover. The flight control equipment monitors these conditions in parallel, and when any condition is met, a recording operation is immediately performed, storing data including position, altitude, speed, attitude, signal strength, and trigger reason identifier as a feature waypoint in non-volatile memory. Optionally, a strategy based on intelligent extraction of flight status can also be used. First, the flight control equipment monitors not only attitude and mode but also GPS positioning quality (such as HDOP value). When a sudden increase in the GPS HDOP value exceeding 2.0 is detected, indicating a decline in satellite positioning quality, the last high-precision positioning point is recorded as a feature waypoint. During recording, the current barometer altitude reading and the relative position information calculated by the IMU are also added to provide redundant altitude and position references in the event of subsequent GPS failure.

[0042] S103. When a communication link interruption and satellite navigation signal loss are detected, the multi-source fusion navigation algorithm is invoked to output the pose estimate at the current moment, and the current interference level is determined based on the preset comprehensive signal quality index.

[0043] Communication link interruption refers to the complete cessation of two-way data communication between the UAV and the ground station, manifested as the failure to receive heartbeat packets from the other party for a preset duration. Satellite navigation signal loss refers to the UAV's inability to receive a sufficient number of valid satellite signals from GNSS systems such as GPS and BeiDou to complete positioning calculations, or the calculated positioning results having extremely low reliability. Multi-source fusion navigation algorithms combine data from multiple different types of sensors (such as IMU, camera, lidar, magnetometer, barometer, etc.) to obtain more accurate and robust navigation information than a single sensor. The pose estimation value represents the algorithm's output estimation results of the UAV's current position, attitude, and velocity. The preset signal quality comprehensive index is a composite evaluation model used to quantify the degree of interference to communication and navigation signals. It integrates multiple parameters such as data link packet loss rate, RSSI, SNR, heartbeat packet interruption duration, and GPS HDOP value. The interference level represents the classification of the current electromagnetic environment interference intensity based on the evaluation results of this comprehensive index, such as mild, moderate, and severe.

[0044] Specifically, the UAV's communication monitoring module and navigation status monitoring module operate in parallel. When the communication module detects that the data link packet loss rate exceeds a threshold (e.g., 50%) and no heartbeat signal is received from the ground station for several consecutive seconds (e.g., 30 seconds), it determines that the communication link is interrupted. Simultaneously, when the navigation module detects that the number of available satellites is less than four or the position accuracy factor (PDOP) remains above a certain threshold, it determines that the satellite navigation signal is lost. Once both conditions are met, the UAV enters an anti-interference emergency state. At this time, the flight control equipment automatically switches the primary navigation source from GPS to a multi-source fusion navigation algorithm. This algorithm (such as a VIO / LIO algorithm based on extended Kalman filtering or factor graph optimization) takes over the pose calculation task, using data from onboard IMU, camera, and lidar sensors to calculate the trajectory and maintain the UAV's basic spatial awareness capabilities. At the same time, based on the last state before the signal interruption and the characteristics of the interruption (e.g., whether the signal gradually decays or disappears instantly), combined with preset comprehensive signal quality indicators, the current interference situation is quickly assessed and classified, outputting judgments such as localized severe interference and regional suppression interference.

[0045] S104. If the current interference level is determined to be local severe interference and the remaining power of the UAV is greater than the first preset threshold, the UAV is controlled to enter the signal search mode. In the signal search mode, the UAV maintains its positioning based on the multi-source fusion navigation algorithm and performs active signal search flight in the direction of signal enhancement based on the spatial signal strength gradient.

[0046] Localized severe interference refers to an interference characteristic characterized by very high interference signal strength, leading to a complete interruption of communication and navigation. However, its impact range is limited, and the UAV may be able to fly out of the interference area through short-distance maneuvers. The first preset threshold is the minimum battery percentage set to initiate the signal search mode, for example, 60%, ensuring that the UAV has sufficient energy to perform the search and leaving a margin for possible return. The signal search mode is an emergency strategy for proactively restoring communication proposed in this invention, with the goal of prioritizing the attempt to restore the link rather than directly returning to base. The spatial signal strength gradient represents the direction and rate of the fastest change in wireless signal strength in three-dimensional space; it is a three-dimensional vector pointing towards the signal source or the direction of signal enhancement.

[0047] Specifically, after determining the interference level to be localized severe interference in step S103, the remaining battery power of the UAV is checked. When the battery power is higher than a preset first threshold (e.g., 60%), it is considered safe and worthwhile to perform an active search. The flight control device does not perform the traditional return-to-home maneuver but switches to signal search mode. In this mode, the UAV's flight control relies entirely on the pose estimation values ​​provided by the multi-source fusion navigation algorithm to maintain flight attitude and perform trajectory tracking. The UAV first performs preset detection maneuvers (e.g., brief hovering, small-range four-way flight, etc.) while using the RSSI detection function of its communication module to measure the signal strength of multiple neighboring points in space. Then, based on these measurement data, the spatial gradient of the signal strength at the current location is calculated. This gradient vector points to the direction in which the signal is most likely to strengthen. Finally, the flight control device uses this gradient direction as the target heading and controls the UAV to fly a distance in this direction, hoping to fly towards the signal recovery area.

[0048] In some embodiments, the search flight based on spatial signal strength gradient in this step can be implemented in various ways. Optionally, a probe-verify-flight loop strategy can be adopted. Specifically, by controlling the UAV to perform a preset probe maneuver (e.g., performing a small-range multi-point maneuver near the current location), signal strength values ​​at multiple locations in space are obtained. The flight control device uses spatial interpolation or fitting algorithms to calculate the rate of change of signal strength in three-dimensional space, thereby extracting the three-dimensional pointing vector pointing to the direction of the fastest signal enhancement. Next, to prevent the gradient from being a multipath spurious signal formed by reflections from mountains or buildings, the flight control device extracts a local two-dimensional elevation profile line with a preset probe distance in the pre-stored digital elevation model (DEM) along the horizontal projection direction of the three-dimensional pointing vector. By obtaining the elevation data of each sampling point on the profile line and its horizontal distance from the UAV, combined with the UAV's current altitude, the elevation angle of each sampling point relative to the UAV's current position can be calculated using basic trigonometric relationships. Subsequently, all sampling points along the profile line are traversed. If the elevation angle of any sampling point is greater than the sum of the UAV's current flight elevation angle and the preset safety margin (for fault tolerance and obstacle avoidance), this physically indicates that the signal gradient direction is obstructed by terrain. In this case, the flight control equipment determines that the three-dimensional pointing vector is not from a direct signal source, but a multipath pseudo-signal generated by terrain reflection, and then eliminates it, reselecting a suboptimal gradient or expanding the detection. If it is not obstructed, the UAV is controlled to perform active signal search flight along the true signal gradient direction. It is understood that other methods can also be used to achieve signal search flight, such as combining the approximate location information of known ground stations or relay stations for directional search, or using systematic search modes such as spiral expansion when there is no obvious gradient; this is not limited here.

[0049] S105. During active signal search flight, if the communication link is not restored and the preset mode switching conditions are met, the signal search mode is interrupted and the UAV is controlled to switch to safe return mode. The preset mode switching conditions include the search duration exceeding the preset duration or the remaining power dropping to the second preset threshold.

[0050] The preset mode switching conditions refer to a set of predefined rules used to determine when to stop signal searching and switch to return-to-home mode. The preset duration is the maximum allowed execution time set for signal search mode, for example, 120 seconds. The second preset threshold is a power warning line set to ensure safe return-to-home, its value being lower than the first preset threshold, for example, 45%. The safe return-to-home mode is another emergency strategy of this invention, its goal being to safely return using existing path information when communication cannot be restored.

[0051] Specifically, as the drone enters signal search mode, the flight control system starts a timer and continuously monitors the remaining battery power. Throughout the search flight, two main switching conditions are periodically checked: first, whether the duration of the search process has exceeded the preset maximum duration (e.g., 120 seconds); and second, whether the drone's remaining battery power has dropped to a preset second threshold (e.g., 45%). Meeting either condition triggers a mode switch. Upon triggering, the flight control system interrupts the current signal search behavior (e.g., stops flying in the gradient direction and enters a hovering state) and switches the primary emergency strategy from signal search mode to safe return-to-home mode.

[0052] In some embodiments, mode switching in this step can be implemented in various ways. Optionally, dynamic switching conditions based on navigation drift error can be introduced. Specifically, when the multi-source fusion navigation algorithm outputs the pose estimate, it simultaneously outputs a position error covariance matrix representing the uncertainty of the system state. The flight control device extracts the eigenvalues ​​(such as diagonal elements) of this matrix to quantitatively evaluate the real-time cumulative drift error of the UAV at the current moment. At the same time, the flight control device calculates the spatial backtracking distance between the current position and the nearest feature waypoint. To calculate the maximum allowable drift threshold, the flight control device first obtains the preset prior drift rate of the multi-source fusion navigation algorithm in the absence of satellite navigation (i.e., the expected error increment per unit flight distance). Multiplying the spatial backtracking distance by the preset prior drift rate allows for the evaluation of the expected additional drift error that the UAV will incur while flying over this backtracking distance, i.e., the expected backtracking drift increment. To ensure that the total positioning error of the UAV remains within the safe capture radius (i.e., the effective range within which it can be successfully identified and entered) when it returns to the designated waypoint, the flight control system subtracts the expected backtracking drift increment from the safe capture radius to obtain the maximum allowable drift threshold at the current moment. The physical meaning of this threshold is that the UAV's accumulated positioning error before starting the backtracking cannot exceed this margin. During signal search, if the real-time accumulated drift error reaches this maximum allowable drift threshold, it means that if the signal search continues, the total error during subsequent backtracking will exceed the safe capture radius, resulting in failure to successfully enter the waypoint. Therefore, even if the search time and battery conditions are not met, the preset mode switching condition is immediately triggered, interrupting the search and forcibly switching to the safe return-to-home mode. It is understood that other methods can also be used to achieve mode switching; for example, if the UAV detects that it has entered a preset no-fly zone during the search, it will also interrupt the search and switch to the safe return-to-home mode. This is not limited here.

[0053] S106. In safe return mode, control the UAV to fly in the opposite direction along the feature track points according to the pose estimation value to the target feature track point that matches the preset emergency safety route space, and then cut into the preset emergency safety route from the target feature track point to perform return.

[0054] The pose estimation value refers to the UAV's position, attitude, and velocity information continuously provided by the multi-source fusion navigation algorithm in the absence of satellite signals. Feature waypoints are the sequence of critical path points recorded in step S102 and stored on the UAV. A target feature waypoint is a special point in the feature waypoint sequence that is spatially close enough to a preset emergency safety route to serve as a connection point for switching from the original return path to the preset emergency safety route.

[0055] Specifically, in safe return-to-home mode, the flight control equipment first accesses the stored list of feature waypoints and, following a last-in-first-out (LIFO) principle, selects the most recently recorded feature waypoint as the first return-to-home target waypoint. The UAV uses the pose estimation value output by the multi-source fusion navigation algorithm to control itself to fly towards this target waypoint. Upon arrival, it then selects the next feature waypoint as the target, and so on, achieving a reverse flight along the original path. During this reverse flight, the flight control equipment continuously performs spatial matching calculations between the current position and the path data of all preset emergency safety routes. Once it detects that the distance between the current feature waypoint or the next feature waypoint to be flown to and a preset emergency safety route is less than a preset capture radius, it identifies that feature waypoint as the target feature waypoint. At this point, the UAV will no longer continue to retreat along the feature waypoint, but will instead enter the matched preset emergency safety route from this target feature waypoint and perform the subsequent return-to-home task along this preset emergency safety route.

[0056] In some embodiments, additional safety and recovery strategies can be added during the return flight along a preset emergency safety route. Specifically, the preset emergency safety route will pre-mark some special waypoints, such as communication reachable points and emergency landing points. When the UAV flies along the preset emergency safety route and passes through areas marked as communication reachable points, the UAV will be controlled to reduce its flight speed or hover briefly to increase the probability of re-establishing the communication link. Throughout the return flight, the remaining battery power will be continuously monitored. If the battery power is lower than a third preset threshold (e.g., 25%, lower than the second threshold), it indicates that it can no longer support the entire flight. The UAV will then abandon the return to the original destination and instead query the route database for the nearest emergency landing point to the current location. The UAV will then be controlled to fly to the emergency landing point. After reaching the airspace above the emergency landing point, the UAV will use onboard visual sensors or lidar to detect ground flatness and obstacles and perform an autonomous vertical descent.

[0057] In the above embodiments, in the case of interference where both communication and satellite navigation fail for the UAV, the cumulative error and risk of yaw and disorientation caused by blindly returning over long distances are avoided. By introducing an active signal search mode, priority is given to attempting to restore the communication link outside the local interference zone. When the search is unsuccessful or the battery is limited, the UAV uses pre-recorded characteristic waypoints and multi-source fusion navigation to reverse back and enter an emergency safety route, thereby reducing the risk of yaw and disorientation of the UAV in complex electromagnetic environments.

[0058] However, the recording of characteristic waypoints mainly relies on changes in flight attitude and flight mode switching. While this method can capture geometric inflection points and logical nodes of the trajectory, it may be difficult to comprehensively record spatial locations important for communication recovery (such as areas of sudden signal enhancement or edges of terrain obstruction) simply by relying on attitude and mode changes in complex terrain and variable electromagnetic environments. If characteristic waypoints lack correlation with communication environment and terrain features, it may lead to missing better communication reconnection opportunities when retreating in safe return mode.

[0059] Please see Figure 2 This is another flowchart illustrating a method for autonomous return of a drone under interference conditions in an embodiment of this application.

[0060] S201. Obtain the pre-generated preset emergency safety route, and monitor the UAV's flight attitude change data and flight mode switching status in real time when the UAV is performing a flight mission and the satellite navigation signal and communication link are normal.

[0061] Step S201 is similar to step S101, and will not be described again here.

[0062] S202. Monitor the signal-to-noise ratio (SNR) data of the airborne communication module in real time. When the SNR at the current moment is greater than the historical average SNR within a preset time window, and the difference between the SNR at the current moment and the historical average SNR is greater than a preset gain threshold, mark the spatial point corresponding to the current moment as the first candidate track point.

[0063] The signal-to-noise ratio (SNR) data refers to the ratio of the signal power to the noise power demodulated by the communication module. The preset time window refers to a time length used to calculate the historical SNR average, such as the past 10 seconds. The historical SNR average is the arithmetic mean of all SNR sampling points within this time window.

[0064] Specifically, during normal drone flight, the flight control equipment reads the current signal-to-noise ratio (SNR) value from the communication module at a certain frequency (e.g., 1Hz). Simultaneously, it maintains a moving average of the SNR readings over a preset time window (e.g., 10 seconds). At each sampling moment, the current instantaneous SNR is compared to this moving average. If the current SNR is not only greater than the historical average, but the difference also exceeds a preset gain threshold (e.g., 3dB), it is considered that the drone has just entered a signal hotspot area with significantly better signal quality. This sudden increase in signal strength usually means that the drone has flown out of an obstructed area or entered the coverage area of ​​the antenna's main lobe. Such spatial points are helpful for subsequent possible signal tracing; the current position coordinates, signal strength, and other information are recorded as a first-choice candidate waypoint for further filtering in subsequent steps.

[0065] S203. Calculate the terrain obstruction angle between the current location and the known ground station based on the pre-stored digital elevation model on the airborne platform.

[0066] A Digital Elevation Model (DEM) is a digital file that represents the undulating terrain of a landmass using a raster data structure, where the value of each raster cell represents the elevation of that point. A known ground station refers to a communication ground control station or relay station whose geographic coordinates have been pre-entered and stored in the UAV during mission planning. The terrain occlusion angle is a geometric angle used to quantify the degree to which the line of sight between the UAV and the ground station is obstructed by terrain.

[0067] Specifically, the flight control equipment first acquires the UAV's current 3D coordinates (from GPS or a fusion navigation system) and pre-stored ground station 3D coordinates. Then, it constructs a theoretically straight line-of-sight vector between these two points. Next, it queries the onboard DEM database to extract all terrain elevation points along the projection path of this line-of-sight vector, forming a terrain profile. By comparing the height of each point on the line-of-sight path with the height of the corresponding terrain profile, the terrain point that causes the greatest obstruction to the line of sight can be identified. The terrain obstruction angle is the difference between the elevation angle of this maximum obstruction point from the UAV's perspective and the depression angle of the ground station. A positive obstruction angle means the line of sight is blocked by that terrain point, while a negative value or zero means the line of sight is unobstructed.

[0068] S204. At the moment when the terrain obstruction angle changes from positive to negative and crosses zero, mark the corresponding spatial point as the second candidate track point.

[0069] The zero-crossing moment refers to the instant when the continuously changing physical quantity of terrain shading angle changes from a positive value to a negative value.

[0070] Specifically, the flight control equipment continuously tracks the changing trend of the terrain obstruction angle. When it detects that the terrain obstruction angle was positive in the previous calculation cycle (indicating that the line of sight was obstructed), and in the current calculation cycle, the angle becomes negative or zero (indicating that the line of sight has just become unobstructed), it determines that a zero-crossing moment has occurred. The physical location corresponding to this zero-crossing moment is precisely the edge point where the UAV flies out of the terrain radio shadow area. This edge point is a critical location where the communication link state may change abruptly, and the spatial location point of the zero-crossing moment is marked as a second candidate waypoint.

[0071] S205. When the flight attitude change data is greater than the preset attitude change threshold or the flight mode switching state changes, mark the spatial point corresponding to the current moment as the third candidate waypoint.

[0072] S206. Based on the preset spatial resolution, the first candidate track point, the second candidate track point, and the third candidate track point are filtered to obtain the feature track point.

[0073] The preset spatial resolution refers to the minimum three-dimensional Euclidean distance set to ensure sufficient spatial spacing between feature waypoints, such as 50 meters. The purpose of this step is to deduplicate and simplify all candidate waypoints generated in the preceding steps.

[0074] Specifically, during the drone's flight, steps S202, S204, and S205 may generate a large number of dense candidate waypoints in a short period. To save storage space and simplify the backtracking path, these candidate points need to be filtered. Whenever a new candidate waypoint (regardless of its type, first, second, or third) is generated, the spatial Euclidean distance between this new point and all feature waypoints already stored in the final list is calculated. If the distance between the new point and an existing waypoint in the list is less than a preset spatial resolution (e.g., 50 meters), the two points are considered too close and need to be discarded. The principle of discarding is to retain the better point, for example, the point with better signal strength data. If the distance between the new point and all existing waypoints is greater than the preset spatial resolution, it is directly added to the final list. Through this process, the final generated sequence of feature waypoints captures all key events while maintaining appropriate sparsity.

[0075] In some embodiments, candidate waypoints can be filtered in various ways. Optionally, a signal strength-based filtering strategy can be adopted. Specifically, the spatial Euclidean distance between any two candidate waypoints is calculated. If the spatial Euclidean distance is less than a preset spatial resolution, the signal strength data (such as RSSI or SNR) corresponding to these two candidate waypoints are compared. The candidate waypoint with the larger signal strength data is retained as the feature waypoint, and the candidate waypoint with the smaller signal strength data is eliminated. Optionally, a multi-level priority filtering strategy can also be adopted. Specifically, priorities are assigned to different types of candidate waypoints. For example, the second candidate waypoint (point with unobstructed line of sight) has the highest priority, followed by the first candidate waypoint (point with improved signal), and the third candidate waypoint (point of maneuver) has the lowest priority. When two candidate waypoints are too close, the higher-priority point is retained first. If the priorities are the same, filtering is then performed based on secondary indicators such as signal strength.

[0076] S207. When a communication link interruption and satellite navigation signal loss are detected, the multi-source fusion navigation algorithm is invoked to output the pose estimate at the current moment, and the current interference level is determined based on the preset comprehensive signal quality index.

[0077] S208. If the current interference level is determined to be local severe interference and the remaining battery power of the UAV is greater than the first preset threshold, the UAV is controlled to enter the signal search mode. In the signal search mode, the UAV maintains its positioning based on the multi-source fusion navigation algorithm and performs active signal search flight in the direction of signal enhancement based on the spatial signal strength gradient.

[0078] S209. During active signal search flight, if the communication link is not restored and the preset mode switching conditions are met, the signal search mode is interrupted and the UAV is controlled to switch to safe return mode. The preset mode switching conditions include the search duration exceeding the preset duration or the remaining power dropping to the second preset threshold.

[0079] S210. In safe return mode, control the UAV to fly in the opposite direction along the feature track points according to the pose estimation value to the target feature track point that matches the preset emergency safety route space, and then cut into the preset emergency safety route from the target feature track point to perform return.

[0080] Step S207 is similar to step S103, step S208 is similar to step S104, step S209 is similar to step S105, and step S210 is similar to step S106, so they will not be described again here.

[0081] In the above embodiments, the first candidate waypoint is extracted by real-time monitoring of the signal-to-noise ratio data of the communication module, the second candidate waypoint is extracted by calculating the zero-crossing time of the terrain occlusion angle using a digital elevation model, and the third candidate waypoint is extracted by combining flight attitude and mode changes. Finally, the candidate points are filtered and deduplicated based on a preset spatial resolution and signal strength, so that the final generated feature waypoint sequence not only includes the geometric features of the trajectory, but also integrates spatial electromagnetic distribution and terrain occlusion information, thereby providing a more accurate and communication recovery-valuable path navigation foundation for UAV signal search and safe retreat in interference environments.

[0082] The above describes a method for autonomous return of a UAV in an interference environment in the embodiments of this application. The following describes an exemplary flight control device 300 provided in the embodiments of this application.

[0083] Figure 3This is a schematic diagram of an exemplary hardware structure of the flight control device 300 provided in an embodiment of this application. In some embodiments, the flight control device 300 is a computer device. The computer device includes a processor, a memory, and a network interface connected via a system bus. The processor of the computer device provides computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The database of the computer device is used to store data. The network interface of the computer device is used to communicate with other external terminals or servers via a network connection. In some embodiments, the network interface can be a wired network interface; in some embodiments, the network interface can also be a wireless network interface. When the computer program is executed by the processor, it implements a method for autonomous return of a UAV in an interference environment according to an embodiment of this application.

[0084] Those skilled in the art will understand that Figure 3 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0085] In some embodiments of this application, a computer-readable storage medium is also provided, including instructions that, when executed on the flight control device 300, enable the flight control device 300 to perform an autonomous return-to-home method for a UAV in an interference environment, as described in this application.

[0086] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit it. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

[0087] As used in the above embodiments, depending on the context, the term "when..." can be interpreted as meaning "if...", "after...", "in response to determining...", or "in response to detecting...". Similarly, depending on the context, the phrase "when determining..." or "if (the stated condition or event) is interpreted as meaning "if determining...", "in response to determining...", "when (the stated condition or event) is detected", or "in response to detecting (the stated condition or event)".

[0088] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state drive), etc.

[0089] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This program can be stored in a computer-readable storage medium, and when executed, it can include the processes described in the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM or random access memory (RAM), magnetic disks, or optical disks.

Claims

1. A method for autonomous return of a drone in a jammed environment, characterized in that, Applied to flight control equipment, the method includes: The system acquires a pre-generated emergency safety route and monitors the UAV's flight attitude change data and flight mode switching status in real time while the UAV is performing a flight mission and the satellite navigation signal and communication link are normal. When the flight attitude change data is greater than the preset attitude change threshold or the flight mode switching state changes, the current position coordinates and signal strength data are recorded as feature waypoints. When the communication link is interrupted and the satellite navigation signal is lost, the multi-source fusion navigation algorithm is invoked to output the pose estimate at the current moment, and the current interference level is determined based on the preset comprehensive signal quality index. If the current interference level is determined to be local severe interference and the remaining battery power of the UAV is greater than the first preset threshold, the UAV is controlled to enter the signal search mode. In the signal search mode, the UAV maintains its positioning according to the multi-source fusion navigation algorithm and performs active signal search flight in the direction of signal enhancement based on the spatial signal strength gradient. During the active signal search flight, if the communication link is not restored and the preset mode switching conditions are met, the signal search mode is interrupted and the UAV is controlled to switch to the safe return mode. The preset mode switching conditions include the search duration exceeding a preset duration or the remaining battery power dropping to a second preset threshold. In the safe return mode, the UAV is controlled to fly in reverse along the feature waypoints according to the pose estimation value to the target feature waypoint that matches the preset emergency safety route space, and then enters the preset emergency safety route from the target feature waypoint to perform the return.

2. The method according to claim 1, characterized in that, The step of recording the current position coordinates and signal strength data as feature track points when the flight attitude change data exceeds a preset attitude change threshold or the flight mode switching state changes specifically includes: The signal-to-noise ratio (SNR) data of the airborne communication module is monitored in real time. When the SNR at the current moment is greater than the historical average SNR within a preset time window, and the difference between the SNR at the current moment and the historical average SNR is greater than a preset gain threshold, the spatial point corresponding to the current moment is marked as the first candidate track point. Calculate the terrain obstruction angle between the current location and a known ground station based on the airborne pre-stored digital elevation model; At the moment when the terrain obstruction angle crosses zero from positive to negative, the corresponding spatial point is marked as the second candidate waypoint; When the flight attitude change data is greater than the preset attitude change threshold or the flight mode switching state changes, the spatial point corresponding to the current moment is marked as the third candidate waypoint. The first candidate waypoint, the second candidate waypoint, and the third candidate waypoint are filtered based on a preset spatial resolution to obtain the feature waypoint.

3. The method according to claim 2, characterized in that, The step of filtering the first candidate waypoint, the second candidate waypoint, and the third candidate waypoint based on a preset spatial resolution to obtain the feature waypoint specifically includes: Calculate the spatial Euclidean distance between any two candidate waypoints; If the spatial Euclidean distance is less than the preset spatial resolution, compare the signal strength data corresponding to the two candidate waypoints; The candidate track point with the larger signal strength data is retained as the feature track point, and the other candidate track point is eliminated.

4. The method according to claim 1, characterized in that, During the active signal search flight, the method further includes: The position error covariance matrix, representing the positioning uncertainty, is obtained in real time from the output of the multi-source fusion navigation algorithm. The spatial backtracking distance is calculated based on the pose estimate at the current moment and the position coordinates of the nearest feature track point; By combining the safe capture radius of the nearest feature trackpoint with the spatial back-off distance, the maximum allowable drift threshold is calculated; When the real-time cumulative drift error corresponding to the position error covariance matrix reaches the maximum allowable drift threshold, the preset mode switching condition is triggered.

5. The method according to claim 4, characterized in that, The step of calculating the maximum permissible drift threshold by combining the safe capture radius of the nearest feature waypoint with the spatial backoff distance specifically includes: Obtain the preset prior drift rate of the multi-source fusion navigation algorithm in the absence of satellite navigation; Multiplying the spatial back-off distance by the preset prior drift rate yields the expected back-off drift increment required for the UAV to fly over the spatial back-off distance; Subtracting the expected backoff drift increment from the safe capture radius yields the maximum allowable drift threshold at the current moment.

6. The method according to claim 1, characterized in that, The step of performing active signal search flight based on the spatial signal intensity gradient towards the signal enhancement direction specifically includes: By controlling the UAV to perform a preset detection maneuver, the signal strength of multiple points in space is obtained, and the three-dimensional pointing vector of the current spatial signal strength gradient is extracted based on the signal strength of the multiple points in space. In the pre-stored digital elevation model on the airborne system, a local two-dimensional elevation profile line is extracted along the three-dimensional pointing vector, starting from the current position of the UAV and ending at a preset detection distance. Calculate the elevation angle of each sampling point on the local two-dimensional elevation profile line relative to the current position of the UAV; If the elevation angle of any sampling point is greater than the sum of the current flight elevation angle of the UAV and the preset safety margin, it is determined that the three-dimensional pointing vector is physically blocked, and the corresponding spatial signal intensity gradient is marked as a pseudo signal gradient caused by multipath reflection and discarded. The UAV is controlled to perform the active signal search flight along the signal gradient direction after removing the pseudo signal gradient.

7. The method according to claim 1, characterized in that, The preset emergency safety route includes pre-marked communication reachable points and emergency landing points. After the step of slicing into the preset emergency safety route from the target feature track point to execute the return flight, the method further includes: While flying along the preset emergency safety route and passing the communication reachable point, reduce the flight speed to attempt to re-establish the communication link; During flight along the preset emergency safety route, if the remaining battery power is lower than the third preset threshold, the drone is controlled to fly to the nearest emergency landing point and use the onboard visual sensor to detect the ground flatness and perform a vertical safe landing.

8. A flight control device, characterized in that, The flight control device includes: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code including computer instructions, and the one or more processors call the computer instructions to cause the flight control device to perform the method as described in any one of claims 1-7.

9. A computer program product containing instructions, characterized in that, When the computer program product is run on the flight control device, the flight control device performs the method as described in any one of claims 1-7.

10. A computer-readable storage medium comprising instructions, characterized in that, When the instruction is executed on the flight control device, the flight control device performs the method as described in any one of claims 1-7.

Images