A method and system for fully autonomous flight path planning and remote take-off and landing control of unmanned aerial vehicles (UAVs)

Abstract

This invention discloses a fully autonomous flight path planning and remote take-off and landing control method and system for unmanned aerial vehicles (UAVs), relating to the field of UAV control technology. The method includes loading a mission dataset containing a high-precision digital elevation model (DEM), a terrain semantic map, and multiple pre-arrival zones into the airborne firmware; performing terrain profile matching based on the DEM in the mission dataset and the vertical altimetry sequence collected in real time during flight to generate continuous pose estimation results; constructing a three-dimensional navigable airspace model using the pose estimation results and the terrain semantic map, and planning an initial mission flight path within this model; during flight along the initial mission flight path, online correction of the initial mission flight path based on updated pose estimation results to form the actual execution flight path and complete the material delivery; after the material delivery is completed, re-optimizing the return trajectory based on remaining fuel and battery power, real-time wind speed, and terrain undulations along the return path to generate an optimized return flight path.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) control technology, and in particular to a method and system for fully autonomous UAV route planning and remote take-off and landing control. Background Technology

[0002] Unmanned aerial vehicle (UAV) control technology refers to a comprehensive hardware and software collaborative approach that uses an airborne flight control system to perceive, make decisions, and execute the attitude, altitude, speed, heading, and mission behavior of a UAV. Its core components include sensor data fusion, state estimation, trajectory planning, flight control law design, and fault-tolerant mechanisms, aiming to achieve stable flight, precise navigation, and autonomous operation of UAVs in various environments. With increasing demands for intelligence and autonomy, modern UAV control technology has gradually evolved from traditional remote control or pre-programmed modes to a fully autonomous control system that integrates environmental perception, real-time decision-making, and adaptive adjustment. Especially under extreme conditions such as the absence of GNSS or communication interruptions, it relies on terrain prior knowledge, airborne intelligence, and robust control strategies to ensure mission completion capabilities.

[0003] Existing drones generally rely heavily on the Global Navigation Satellite System for positioning. In environments such as canyons, dense forests, electromagnetic interference, or enemy suppression, GNSS signals are easily lost, causing the flight control system to quickly lose position reference and become unable to continue performing tasks or return to base. Furthermore, after the remote control or data transmission link is interrupted, traditional drones can usually only perform simple return-to-home or hovering and waiting, and cannot cope with battlefield or disaster scenarios such as when the original take-off and landing point has been destroyed or is unavailable, lacking true remote take-off and landing capabilities. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a fully autonomous flight path planning and remote take-off and landing control method for unmanned aerial vehicles (UAVs) to solve the problem that existing UAVs generally rely heavily on the Global Navigation Satellite System (GNSS) for positioning. In environments such as canyons, dense forests, electromagnetic interference, or enemy suppression, GNSS signals are easily lost, causing the flight control system to quickly lose its position reference and become unable to continue the mission or return to base. Furthermore, after the remote control or data transmission link is interrupted, traditional UAVs can usually only perform simple return-to-home or hovering wait, and cannot cope with battlefield or disaster scenarios such as when the original take-off and landing point has been destroyed or is unavailable, lacking true remote take-off and landing capabilities.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, the present invention provides a method for fully autonomous flight path planning and remote take-off and landing control of unmanned aerial vehicles (UAVs), comprising: Load the mission dataset, which includes a high-precision digital elevation model, terrain semantic map, and multiple pre-equipment drop zones, into the airborne firmware; Based on the digital elevation model in the task dataset and the vertical altimetry sequence collected in real time during flight, terrain profile matching is performed to generate continuous pose estimation results. The pose estimation results and the terrain semantic map are used to construct a three-dimensional navigable airspace model, and the initial mission route is planned within the navigable airspace model. During flight along the initial mission route, the initial mission route is corrected online based on the updated pose estimation results to form the actual execution route and complete the delivery of supplies. After the supplies are delivered, the return trajectory is re-optimized based on the remaining fuel and power, real-time wind speed, and terrain undulations of the return route, and an optimized return route is generated. When multiple communication and navigation failures are detected, the minimum survival return state machine is activated, the flight is guided according to the optimized return route, and the landing target point is selected based on the pre-equipped landing area and real-time sensed ground environmental parameters to complete the autonomous landing.

[0007] As a preferred embodiment of the fully autonomous flight path planning and remote take-off and landing control method for unmanned aerial vehicles (UAVs) described in this invention, the specific steps for performing terrain profile matching and generating continuous pose estimation results are as follows: Vertical altitude measurements are acquired at a fixed sampling period during flight to form a real-time altitude sequence. ; Extract reference altitude sequences of corresponding lengths from the digital elevation model in the mission dataset along the predicted flight path. ,in Indicates the horizontal position coordinates of the candidate; For each candidate position Calculate the sequence and Dynamic time-warped distance between ; Select to make Minimum position As the current horizontal positioning result; Integrate the output of the inertial measurement unit with the barometer data, and combine them. Constructing a six-DOF pose estimation method; Among them, dynamic time warping distance Calculate using the following formula: ; in, This represents an alignment path that satisfies monotonicity and boundary constraints. For the real-time altimeter sequence One value; In the candidate position The first point extracted from the digital elevation model A reference height value.

[0008] As a preferred embodiment of the fully autonomous flight path planning and remote take-off and landing control method for unmanned aerial vehicles (UAVs) described in this invention, the specific steps of constructing a three-dimensional navigable airspace model using pose estimation results and terrain semantic maps, and planning an initial mission flight path within this navigable airspace model, are as follows: Identify impassable areas from topographic semantic maps, including areas with excessive slope, dense forests, and waterways; Based on the digital elevation model, a safe clearance height is superimposed above each geographic grid point to generate the upper limit of the flight altitude; Define the flyable voxel space as the set of all three-dimensional grid points that are not marked as impassable and whose height is below the flyable upper limit; In the flyable voxel space, the initial mission flight path is generated using a three-dimensional path search algorithm, with the initial pose as the starting point and the material delivery point as the ending point. To improve the communication and perception reliability of flight routes in complex mountainous environments, a terrain occlusion cost function is introduced to participate in path evaluation; Among them, the terrain occlusion cost function Calculate using the following formula: ; in, This represents the total number of sampling points along the line of sight. Indicates the first Normalized position of each sampling point; This is the height of the digital elevation model at that point; The interpolated height of the straight line connecting the current position of the drone and the reference point at that position; This is an indicator function that takes the value 1 if the condition is true, and 0 otherwise.

[0009] As a preferred embodiment of the fully autonomous flight path planning and remote take-off and landing control method for unmanned aerial vehicles (UAVs) described in this invention, the step of re-optimizing the return trajectory by combining remaining fuel and battery power, real-time wind speed, and terrain undulations along the return path to generate an optimized return route includes the following specific steps: Obtain the current remaining fuel and battery level, and the corresponding maximum flight energy. For each alternate landing zone, calculate the candidate return route from the current location to that alternate landing zone; The total energy consumption is calculated by integrating along each candidate path; Select paths whose total energy consumption does not exceed the safety margin limit; Choose the route with the lowest energy consumption that meets the conditions as the optimized return route; Among them, total energy consumption of the path Calculate using the following formula: ; in, This is the basic energy consumption coefficient per unit distance during level flight; The climbing energy consumption gain coefficient; This is the gain coefficient for headwind energy consumption; For the path along the arc length The rate of change of altitude reflects the terrain slope; This is the local wind speed vector; This is the tangential unit vector along the path.

[0010] As a preferred embodiment of the fully autonomous flight path planning and remote take-off and landing control method for unmanned aerial vehicles (UAVs) described in this invention, the method involves selecting a landing target point based on real-time sensed ground environmental parameters to complete autonomous landing. The specific steps are as follows: Activate the airborne scanning sensor to acquire local ground point cloud data; Calculate the average slope of the area. Surface roughness and obstacle density ; If all three indicators do not exceed their respective safety thresholds, the area is confirmed as a usable landing zone. Of all available landing zones, the one closest along the return route is selected as the final landing target point; Execute the descent and ground contact buffer control process; Among them, the comprehensive landing suitability score Calculate using the following formula: ; in, , , These are positive weighting coefficients; The average slope; The standard deviation of surface elevation characterizes surface roughness; The number of obstacles per unit area; , , These are the corresponding security thresholds.

[0011] As a preferred embodiment of the fully autonomous flight path planning and remote take-off and landing control method for unmanned aerial vehicles (UAVs) described in this invention, the execution of the descent and ground contact buffer control process specifically includes: Hovering above the final landing target point; It descends at a constant rate while monitoring changes in the altimeter radar echo; When the rate of change of the echo distance exceeds the trigger threshold, it is determined that the device is about to touch the ground. Immediately reduce rotor thrust and activate the mechanical buffer device; After touching the ground, power is cut off and the device enters a dormant state.

[0012] As a preferred embodiment of the fully autonomous flight path planning and remote take-off and landing control method for unmanned aerial vehicles (UAVs) described in this invention, the minimum survival return state machine is activated when three or more of the following events—loss of global navigation satellite system signal, interruption of remote control signal, interruption of image transmission signal, and interruption of data transmission signal—continue for more than a set duration: after activation, non-essential functional modules are shut down, and only the inertial measurement unit, barometer, altimeter, and flight control core loop are retained. The optimized return flight path height profile is forcibly locked for terrain-following flight, ensuring that the return mission can still be completed under extreme communication and navigation failure conditions.

[0013] Secondly, this invention provides a fully autonomous flight path planning and remote take-off and landing control system for unmanned aerial vehicles (UAVs), comprising: The module includes a mission data loading module, a terrain matching and positioning module, a passable airspace construction module, an online flight path correction module, an energy-sensing return module, and a flexible landing decision module. The task data loading module is used to load the high-precision digital elevation model, terrain semantic map and task dataset of multiple pre-equipment landing zones into the airborne firmware during the task initialization phase. The terrain matching and positioning module is used to collect vertical altimetry sequences during flight, combine them with digital elevation models to perform dynamic time warping matching of terrain profiles, generate continuous pose estimation results, and support autonomous navigation in GNSS-free environments. The passable airspace construction module is used to identify impassable areas based on pose estimation results and terrain semantic map, generate a three-dimensional flyable voxel space by superimposing safe clearance height, plan the initial mission route in the space, and introduce a terrain occlusion cost function to optimize communication and perception reliability. The online flight path correction module is used to determine the lateral deviation based on the updated pose estimation results during flight along the initial mission flight path. When the deviation exceeds the tolerance, it triggers local replanning, generates the actual execution flight path, and guides the completion of material delivery. The energy sensing return module is used to optimize the return trajectory after the material delivery is completed by combining the remaining fuel and electricity, real-time wind speed and terrain undulation of the return path, and selecting the optimal return route that meets the energy safety margin. The elastic landing decision module is used to activate the minimum survival return state machine when multiple communication and navigation failures are detected, guide the flight according to the optimized return route, and calculate the comprehensive landing suitability score by integrating real-time sensed slope, roughness and obstacle density when approaching the pre-arrival landing area, select the optimal landing target point and execute soft landing control.

[0014] Thirdly, the present invention provides a computer device including a memory and a processor, wherein the memory stores a computer program, wherein when the computer program is executed by the processor, it implements any step of the fully autonomous flight path planning and remote take-off and landing control method for unmanned aerial vehicles as described in the first aspect of the present invention.

[0015] Fourthly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the fully autonomous flight path planning and remote take-off and landing control method for unmanned aerial vehicles as described in the first aspect of the present invention.

[0016] The beneficial effects of this invention are as follows: By integrating high-precision terrain priors with airborne real-time perception, under the extreme conditions of no GNSS, no communication, no remote control, and no image transmission, the UAV can achieve fully autonomous mission execution and safe return from any starting point to any destination. The proposed terrain fingerprint matching positioning, terrain occlusion perception route planning, terrain-wind field coupled energy consumption model, and elastic landing decision mechanism improve the system's navigation robustness, path safety, energy utilization efficiency, and landing reliability in complex mountainous, battlefield, or disaster environments. At the same time, the minimum survival return state machine ensures that the system still has basic return capability under multiple faults. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart of the fully autonomous flight path planning and remote take-off and landing control method for unmanned aerial vehicles in Example 1.

[0019] Figure 2 This is a schematic diagram of the fully autonomous flight path planning and remote take-off and landing control system for UAVs in Example 1. Detailed Implementation

[0020] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0021] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0022] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0023] Example 1, referring to Figure 1 and Figure 2 This is one embodiment of the present invention, which provides a method for fully autonomous flight path planning and remote take-off and landing control of unmanned aerial vehicles (UAVs), comprising the following steps: S1. Load the task dataset containing a high-precision digital elevation model, terrain semantic map and multiple pre-equipment drop zones into the airborne firmware.

[0024] Furthermore, the mission dataset is generated by the mission command terminal during the ground mission planning phase and written to the non-volatile storage area of ​​the UAV's onboard firmware via an encrypted secure channel; the spatial resolution of the high-precision digital elevation model reaches ten meters or higher, covering the entire flight airspace of the mission area; the terrain semantic map labels the surface type of each geographic unit in raster form, including passable flat land, areas with excessive slope, dense forest areas, water areas, and artificial building areas; multiple pre-selected landing zones are candidate landing points that meet the minimum area, maximum slope, electromagnetic environment, and line-of-sight conditions, and each alternate landing zone is associated with its geographic coordinates, surface attributes, and priority weight; after the UAV completes its power-on self-test, the firmware automatically verifies the integrity and digital signature of the mission dataset, and after successful verification, loads the data into the running memory for subsequent navigation and decision-making modules to call.

[0025] It should be noted that the structured loading and security verification mechanism of the aforementioned task dataset ensures that the UAV has complete prior environmental knowledge and legitimate mission authorization before takeoff, avoiding navigation failures or misoperations due to missing, tampered, or inconsistent data. The collaborative use of high-resolution digital elevation models and refined terrain semantic maps provides a reliable foundation for subsequent GNSS-free positioning, navigable airspace construction, and landing decisions, improving the system's mission preparation completeness and execution certainty in unknown or adversarial environments.

[0026] S2. Based on the digital elevation model in the mission dataset and the vertical altimetry sequence collected in real time during flight, perform terrain profile matching to generate continuous pose estimation results.

[0027] Furthermore, vertical altitude measurements are acquired at a fixed sampling period during flight to form a real-time altitude sequence. ; Extract reference altitude sequences of corresponding lengths from the digital elevation model in the mission dataset along the predicted flight path. ,in Indicates the horizontal position coordinates of the candidate; For each candidate position Calculate the sequence and Dynamic time-warped distance between ; Select to make Minimum position As the current horizontal positioning result; Integrate the output of the inertial measurement unit with the barometer data, and combine them. Constructing a six-DOF pose estimation method; Among them, dynamic time warping distance Calculate using the following formula: ; in, This represents an alignment path that satisfies monotonicity and boundary constraints. For the real-time altimeter sequence One value; In the candidate position The first point extracted from the digital elevation model A reference height value.

[0028] It should be noted that the nonlinear matching between the altimeter sequence and the terrain profile is achieved through the dynamic time warping algorithm, which effectively overcomes the positioning drift problem caused by flight speed fluctuations, asynchronous sampling, or terrain repetition. This method does not rely on satellite signals and can achieve sub-hundred-meter level continuous pose estimation using only airborne altimeter sensors and pre-stored terrain data. It provides underlying navigation support for fully autonomous flight in four-segment scenarios and solves the inherent defects of traditional inertial navigation such as long-term drift and visual navigation being susceptible to light interference.

[0029] S3. Construct a three-dimensional accessible airspace model using the pose estimation results and terrain semantic map, and plan the initial mission route within this accessible airspace model.

[0030] Furthermore, impassable areas can be identified from the topographic semantic map, including areas with excessive slope, dense forests, and water bodies; Based on the digital elevation model, a safe clearance height is superimposed above each geographic grid point to generate the upper limit of the flight altitude; Define the flyable voxel space as the set of all three-dimensional grid points that are not marked as impassable and whose height is below the flyable limit; In the flyable voxel space, the initial mission flight path is generated using a three-dimensional path search algorithm, with the initial pose as the starting point and the material delivery point as the ending point. To improve the communication and perception reliability of flight routes in complex mountainous environments, a terrain occlusion cost function is introduced to participate in path evaluation; Among them, the terrain occlusion cost function Calculate using the following formula: ; in, This represents the total number of sampling points along the line of sight. Indicates the first Normalized position of each sampling point; This is the height of the digital elevation model at that point; The interpolated height of the straight line connecting the current position of the drone and the reference point at that position; This is an indicator function that takes the value 1 if the condition is true, and 0 otherwise.

[0031] It should be noted that by integrating terrain semantic information with digital elevation models to construct a three-dimensional navigable airspace and introducing a terrain occlusion cost function for path evaluation, the planned initial mission route not only avoids physical obstacles but also actively avoids communication blind spots and perception dead zones. This strategy improves the connectivity and safety of UAVs in complex mountainous, canyon, or urban ruin environments, breaking through the limitations of traditional path planning that only focuses on geometric obstacle avoidance.

[0032] S4. During flight along the initial mission route, the initial mission route is corrected online based on the updated pose estimation results to form the actual execution route and complete the delivery of supplies.

[0033] Furthermore, the online correction process is executed with a fixed control cycle: First, the latest pose estimate output by the terrain matching and positioning module is obtained at the current moment; then, the lateral deviation between the projection point of this pose on the horizontal plane and the nearest waypoint of the initial mission route is calculated; if the lateral deviation is greater than a preset tolerance threshold, the current pose is used as a new starting point, and a short-range correction segment connecting the current position and the original mission endpoint is generated using a fast replanning algorithm within the locally traversable airspace; the short-range correction segment is smoothly spliced ​​with the unflyed part of the initial mission route to form the actual execution route; the flight controller switches to tracking the actual execution route and continuously monitors pose updates; when the UAV arrives within the geographical coordinates of the material delivery point and the altitude is below the delivery threshold, the material release mechanism is triggered to complete the delivery action, and the delivery completion status is recorded and the return preparation process is initiated.

[0034] It should be noted that the online flight path correction mechanism triggered by real-time pose updates enables a closed-loop response to the accumulation of positioning errors and dynamic changes in the environment. Through local replanning and smooth stitching, the continuity of the flight trajectory and control stability are ensured, while ensuring that supplies are accurately delivered to the target area. This process is completed autonomously on the airborne end without external intervention, which enhances the robustness and mission success rate of the system in long-endurance and long-distance missions.

[0035] S5. After the supplies are delivered, the return trajectory is re-optimized based on the remaining fuel and power, real-time wind speed, and terrain undulations along the return route, and an optimized return route is generated.

[0036] Furthermore, obtain the current remaining fuel and battery power, corresponding to the maximum flight energy limit; For each alternate landing zone, calculate the candidate return route from the current location to that alternate landing zone; The total energy consumption is calculated by integrating along each candidate path; Select paths whose total energy consumption does not exceed the safety margin limit; Choose the route with the lowest energy consumption that meets the conditions as the optimized return route; Among them, total energy consumption of the path Calculate using the following formula: ; in, This is the basic energy consumption coefficient per unit distance during level flight; The climbing energy consumption gain coefficient; This is the gain coefficient for headwind energy consumption; For the path along the arc length The rate of change of altitude reflects the terrain slope; This is the local wind speed vector; This is the tangential unit vector along the path.

[0037] It should be noted that the terrain-wind field coupled energy consumption model adopted explicitly incorporates climb power consumption and headwind resistance into the return-to-home decision, making the energy estimation closer to the actual flight physics characteristics. Based on this model, a return-to-home path that meets the safety margin is selected, effectively preventing mid-flight crashes due to insufficient power, and is especially suitable for high-energy-consumption scenarios such as plateaus and mountains. This method realizes intelligent survival return-to-home under energy-constrained conditions and is a key support for the four-segment availability capability.

[0038] S6. When multiple communication and navigation failures are detected, the minimum survival return state machine is activated, the flight is guided according to the optimized return route, and the landing target point is selected based on the pre-equipped landing area and real-time sensed ground environmental parameters to complete the autonomous landing.

[0039] Furthermore, the airborne scanning sensor is activated to acquire local ground point clouds; Calculate the average slope of the area. Surface roughness and obstacle density ; If all three indicators do not exceed their respective safety thresholds, the area is confirmed as a usable landing zone. Of all available landing zones, the one closest along the return route is selected as the final landing target point; Execute the descent and ground contact buffer control process; Among them, the comprehensive landing suitability score Calculate using the following formula: ; in, , , These are positive weighting coefficients; The average slope; The standard deviation of surface elevation characterizes surface roughness; The number of obstacles per unit area; , , These are the corresponding security thresholds; The descent and ground contact buffer control procedure includes: Hovering above the final landing target point; It descends at a constant rate while monitoring changes in the altimeter radar echo; When the rate of change of the echo distance exceeds the trigger threshold, it is determined that the device is about to touch the ground. Immediately reduce rotor thrust and activate the mechanical buffer device; After touching the ground, power is cut off, and the system enters a dormant state. The minimum survivability return state machine is activated when three or more of the following are detected for a continuous period of time: loss of Global Navigation Satellite System signal, interruption of remote control signal, interruption of image transmission signal, and interruption of data transmission signal. After activation, non-essential functional modules are shut down, and only the inertial measurement unit, barometer, altimeter radar, and flight control core loop are retained. It also forcibly locks the altitude profile of the optimized return route for terrain-following flight, ensuring that the return mission can still be completed under extreme communication and navigation failure conditions.

[0040] It should be noted that the combination of the minimum survivability return state machine and the multi-dimensional landing suitability scoring mechanism enables the UAV to autonomously select the safest landing point and perform a soft landing even under extreme failure conditions; by shutting down unnecessary modules while retaining core functions, the available flight time is extended to the maximum extent; and the ground contact criterion based on the rate of change of altimeter radar echo does not depend on absolute altitude, adapts to unknown terrain deviations, and improves landing reliability; the overall design ensures that even under severe conditions of complete loss of communication, no GNSS, and partial failure of multiple sensors, the system can still complete a fully autonomous closed-loop operation of flying out, returning, and landing stably, meeting the core requirements of high-reliability military equipment.

[0041] This embodiment also provides a fully autonomous flight path planning and remote take-off and landing control system for unmanned aerial vehicles (UAVs), including: The module includes a mission data loading module, a terrain matching and positioning module, a passable airspace construction module, an online flight path correction module, an energy-sensing return module, and a flexible landing decision module. The mission data loading module is used to load the high-precision digital elevation model, terrain semantic map and mission datasets of multiple pre-equipment landing zones into the airborne firmware during the mission initialization phase. The terrain matching and positioning module is used to collect vertical altimetry sequences during flight, combine them with digital elevation models to perform dynamic time warping matching of terrain profiles, generate continuous pose estimation results, and support autonomous navigation in GNSS-free environments. The passable airspace construction module is used to identify impassable areas based on pose estimation results and terrain semantic maps, overlay safe clearance height to generate a three-dimensional flyable voxel space, and plan the initial mission route in this space. At the same time, a terrain occlusion cost function is introduced to optimize communication and perception reliability. The online flight path correction module is used to determine the lateral deviation based on the updated pose estimation results during flight along the initial mission flight path. When the deviation exceeds the tolerance, it triggers local replanning, generates the actual execution flight path, and guides the completion of the material delivery. The energy sensing return module is used to optimize the return trajectory after the material delivery is completed by combining the remaining fuel and electricity, real-time wind speed and terrain undulation of the return path, and selecting the optimal return route that meets the energy safety margin. The flexible landing decision module is used to activate the minimum survival return state machine when multiple communication and navigation failures are detected. It guides the flight based on the optimized return route and, when approaching the pre-arrival landing area, integrates real-time sensed slope, roughness and obstacle density to calculate a comprehensive landing suitability score, select the optimal landing target point and execute soft landing control.

[0042] This embodiment also provides a computer device applicable to the fully autonomous flight path planning and remote take-off and landing control method for unmanned aerial vehicles (UAVs), including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to realize the fully autonomous flight path planning and remote take-off and landing control method for UAVs as proposed in the above embodiment.

[0043] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0044] This embodiment also provides a storage medium storing a computer program. When executed by a processor, the program implements the fully autonomous flight path planning and remote take-off and landing control method for unmanned aerial vehicles (UAVs) as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0045] In summary, this invention integrates high-precision terrain prior knowledge with airborne real-time perception, enabling UAVs to perform fully autonomous missions and return safely from any starting point to any destination under extreme conditions of no GNSS, no communication, no remote control, and no image transmission. The proposed terrain fingerprint matching positioning, terrain occlusion perception route planning, terrain-wind field coupled energy consumption model, and elastic landing decision mechanism improve the system's navigation robustness, path safety, energy utilization efficiency, and landing reliability in complex mountainous, battlefield, or disaster environments. At the same time, the minimum survivability return state machine ensures that the system still has basic return capability under multiple failures.

[0046] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for fully autonomous flight path planning and remote take-off and landing control of unmanned aerial vehicles (UAVs), characterized in that: include: Load the mission dataset, which includes a high-precision digital elevation model, terrain semantic map, and multiple pre-equipment drop zones, into the airborne firmware; Based on the digital elevation model in the task dataset and the vertical altimetry sequence collected in real time during flight, terrain profile matching is performed to generate continuous pose estimation results. The pose estimation results and the terrain semantic map are used to construct a three-dimensional navigable airspace model, and the initial mission route is planned within the navigable airspace model. During flight along the initial mission route, the initial mission route is corrected online based on the updated pose estimation results to form the actual execution route and complete the delivery of supplies. After the supplies are delivered, the return trajectory is re-optimized based on the remaining fuel and power, real-time wind speed, and terrain undulations of the return route, and an optimized return route is generated. When multiple communication and navigation failures are detected, the minimum survival return state machine is activated, the flight is guided according to the optimized return route, and the landing target point is selected based on the pre-equipped landing area and real-time sensed ground environmental parameters to complete the autonomous landing.

2. The fully autonomous flight path planning and remote take-off and landing control method for unmanned aerial vehicles as described in claim 1, characterized in that: The specific steps for performing terrain profile matching and generating continuous pose estimation results are as follows: Vertical altitude measurements are acquired at a fixed sampling period during flight to form a real-time altitude sequence. ; Extract reference altitude sequences of corresponding lengths from the digital elevation model in the mission dataset along the predicted flight path. ,in Indicates the horizontal position coordinates of the candidate; For each candidate position Calculate the sequence and Dynamic time-warped distance between ; Select to make Minimum position As the current horizontal positioning result; Integrate the output of the inertial measurement unit with the barometer data, and combine them. Constructing a six-DOF pose estimation method; Among them, dynamic time warping distance Calculate using the following formula: ； in, This represents an alignment path that satisfies monotonicity and boundary constraints. For the real-time altimeter sequence One value; In the candidate position The first point extracted from the digital elevation model A reference height value.

3. The fully autonomous flight path planning and remote take-off and landing control method for unmanned aerial vehicles as described in claim 2, characterized in that: The specific steps for constructing a 3D navigable airspace model using pose estimation results and terrain semantic maps, and planning an initial mission route within this navigable airspace model, are as follows: Identify impassable areas from topographic semantic maps, including areas with excessive slope, dense forests, and waterways; Based on the digital elevation model, a safe clearance height is superimposed above each geographic grid point to generate the upper limit of the flight altitude; Define the flyable voxel space as the set of all three-dimensional grid points that are not marked as impassable and whose height is below the flyable upper limit; In the flyable voxel space, the initial mission flight path is generated using a three-dimensional path search algorithm, with the initial pose as the starting point and the material delivery point as the ending point. To improve the communication and perception reliability of flight routes in complex mountainous environments, a terrain occlusion cost function is introduced to participate in path evaluation; Among them, the terrain occlusion cost function Calculate using the following formula: ; in, This represents the total number of sampling points along the line of sight. Indicates the first Normalized position of each sampling point; This is the height of the digital elevation model at that point; The interpolated height of the straight line connecting the current position of the drone and the reference point at that position; This is an indicator function that takes the value 1 if the condition is true, and 0 otherwise.

4. The fully autonomous flight path planning and remote take-off and landing control method for unmanned aerial vehicles as described in claim 3, characterized in that: The process of combining remaining fuel and battery power, real-time wind speed, and terrain undulations along the return route to re-optimize the return trajectory and generate an optimized return route involves the following steps: Obtain the current remaining fuel and battery level, and the corresponding maximum flight energy. For each alternate landing zone, calculate the candidate return route from the current location to that alternate landing zone; The total energy consumption is calculated by integrating along each candidate path; Select paths whose total energy consumption does not exceed the safety margin limit; Choose the route with the lowest energy consumption that meets the conditions as the optimized return route; Among them, total energy consumption of the path Calculate using the following formula: ; in, This is the basic energy consumption coefficient per unit distance during level flight; The climbing energy consumption gain coefficient; This is the gain coefficient for headwind energy consumption; For the path along the arc length The rate of change of altitude reflects the terrain slope; This is the local wind speed vector; This is the tangential unit vector along the path.

5. The fully autonomous flight path planning and remote take-off and landing control method for unmanned aerial vehicles as described in claim 4, characterized in that: Upon approaching the pre-arrival landing zone, the system selects a landing target point based on real-time sensed ground environmental parameters to complete the autonomous landing. The specific steps are as follows: Activate the airborne scanning sensor to acquire local ground point cloud data; Calculate the average slope of the area. Surface roughness and obstacle density ; If all three indicators do not exceed their respective safety thresholds, the area is confirmed as a usable landing zone. Of all available landing zones, the one closest along the return route is selected as the final landing target point; Execute the descent and ground contact buffer control process; Among them, the comprehensive landing suitability score Calculate using the following formula: ; in, , , These are positive weighting coefficients; The average slope; The standard deviation of surface elevation characterizes surface roughness; The number of obstacles per unit area; , , These are the corresponding security thresholds.

6. The fully autonomous flight path planning and remote take-off and landing control method for unmanned aerial vehicles as described in claim 5, characterized in that: The execution of the descent and ground contact buffer control process specifically includes: Hovering above the final landing target point; It descends at a constant rate while monitoring changes in the altimeter radar echo; When the rate of change of the echo distance exceeds the trigger threshold, it is determined that the device is about to touch the ground. Immediately reduce rotor thrust and activate the mechanical buffer device; After touching the ground, power is cut off and the device enters a dormant state.

7. The fully autonomous flight path planning and remote take-off and landing control method for unmanned aerial vehicles as described in claim 6, characterized in that: The minimum survival return state machine is activated when three or more of the following are detected for a continuous period of time: loss of global navigation satellite system signal, interruption of remote control signal, interruption of image transmission signal, and interruption of data transmission signal: all for a set duration. After activation, non-essential functional modules are shut down, and only the inertial measurement unit, barometer, altimeter radar, and flight control core loop are retained. The optimized return route altitude profile is forcibly locked for terrain-following flight to ensure that the return mission can still be completed under extreme communication and navigation failure conditions.

8. A fully autonomous flight path planning and remote take-off and landing control system for unmanned aerial vehicles (UAVs), based on the fully autonomous flight path planning and remote take-off and landing control method for UAVs as described in any one of claims 1 to 7, characterized in that: include: The module includes a mission data loading module, a terrain matching and positioning module, a passable airspace construction module, an online flight path correction module, an energy-sensing return module, and a flexible landing decision module. The task data loading module is used to load the high-precision digital elevation model, terrain semantic map and task dataset of multiple pre-equipment landing zones into the airborne firmware during the task initialization phase. The terrain matching and positioning module is used to collect vertical altimetry sequences during flight, combine them with digital elevation models to perform dynamic time warping matching of terrain profiles, generate continuous pose estimation results, and support autonomous navigation in GNSS-free environments. The passable airspace construction module is used to identify impassable areas based on pose estimation results and terrain semantic map, generate a three-dimensional flyable voxel space by superimposing safe clearance height, plan the initial mission route in the space, and introduce a terrain occlusion cost function to optimize communication and perception reliability. The online flight path correction module is used to determine the lateral deviation based on the updated pose estimation results during flight along the initial mission flight path. When the deviation exceeds the tolerance, it triggers local replanning, generates the actual execution flight path, and guides the completion of material delivery. The energy sensing return module is used to optimize the return trajectory after the material delivery is completed by combining the remaining fuel and electricity, real-time wind speed and terrain undulation of the return path, and selecting the optimal return route that meets the energy safety margin. The elastic landing decision module is used to activate the minimum survival return state machine when multiple communication and navigation failures are detected, guide the flight according to the optimized return route, and calculate the comprehensive landing suitability score by integrating real-time sensed slope, roughness and obstacle density when approaching the pre-arrival landing area, select the optimal landing target point and execute soft landing control.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that: When the processor executes the computer program, it implements the steps of the fully autonomous flight path planning and remote take-off and landing control method for unmanned aerial vehicles as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by the processor, it implements the steps of the fully autonomous flight path planning and remote take-off and landing control method for unmanned aerial vehicles as described in any one of claims 1 to 7.

Images