Drone homing method and drone hangar

Abstract

The application relates to the technical field of low-altitude aircrafts, and discloses a method for returning an unmanned aerial vehicle (UAV) to a position and a UAV garage, the method comprising the following steps: acquiring a landing position coordinate of the UAV; when a position deviation between the landing position coordinate and a target position coordinate is greater than a first preset deviation, acquiring the weight of the UAV; determining a basic thrust of a mechanical arm according to the weight of the UAV; dividing the position deviation into multiple intervals, determining an actual thrust of the mechanical arm based on the basic thrust according to the interval where the position deviation is located; and controlling the mechanical arm to apply the determined actual thrust to the UAV, so as to move the UAV to the target position. After the UAV lands, the mechanical arm outputs a matching force according to the actual weight and the real-time position deviation, effectively avoids the problems of position deviation and attitude deviation of the UAV caused by the mismatching of the size and direction of the force, and significantly improves the returning accuracy and stability of the UAV.

Description

Technical Field

[0001] This invention relates to the field of low-altitude aircraft technology, specifically to a method for locating unmanned aerial vehicles (UAVs) and a UAV hangar. Background Technology

[0002] With the increasing demand for automated drone operations, precise drone docking and recovery have become key technical challenges in the industry. Most drone landing and recovery devices in related technologies employ fixed structural supports or passive support methods. After the drone enters the docking area, they rely heavily on robotic arms for simple clamping or positioning, achieving only basic positional fixation. In actual landing processes, landing positions vary across different operational scenarios, and the drones themselves exhibit different states. The robotic arm struggles to output appropriate support forces based on the actual landing conditions. When the drone lands at the designated location, the magnitude and direction of the force output by the robotic arm may not match the landing state, easily leading to problems such as positional deviation, slippage, or even tipping over. This makes it difficult to stably dock at the target position and fails to meet the requirements for high-precision, high-stability automated landing and recovery. Summary of the Invention

[0003] This invention provides a method for drone repositioning and a drone storage system to solve the problem that when a drone lands at a designated location, the robotic arm of the drone landing and recovery device cannot output an appropriate supporting force, making it difficult to achieve stable and precise automated landing and recovery.

[0004] In a first aspect, the present invention provides a method for returning a drone to its original location, comprising: Obtain the landing location coordinates of the drone; If the positional deviation between the landing position coordinates and the target position coordinates is greater than a first preset deviation, then the weight of the drone is obtained; Determine the basic thrust of the robotic arm based on the weight of the drone; The position deviation is divided into multiple intervals, and the actual thrust of the robotic arm is determined based on the base thrust according to the interval where the position deviation is located. The robotic arm is controlled to apply a determined actual thrust to the drone, moving the drone to the target position.

[0005] Beneficial effects: By acquiring the landing position coordinates of the UAV and further acquiring the UAV's weight when the position deviation exceeds the limit, the basic thrust of the robotic arm is determined based on the weight. Then, the basic thrust is adaptively adjusted in combination with the position deviation range to obtain a suitable actual thrust to control the robotic arm's movements. After the UAV lands, it can output a matching force based on its actual weight and real-time position deviation, effectively avoiding problems such as UAV position deviation, attitude deviation, slippage, or even tipping caused by mismatch between the magnitude and orientation of the force. This significantly improves the accuracy and stability of the UAV's return to its original position, reliably achieving precise adjustment and stable docking of the UAV to the target location, and meeting the requirements of automated, high-precision landing and recovery.

[0006] In one alternative implementation, determining the basic thrust of the robotic arm based on the weight of the drone includes: When the weight of the drone is less than the first preset weight, the basic thrust of the robotic arm is determined to be F1; When the weight of the drone is equal to or greater than the first preset weight and less than the second preset weight, the basic thrust of the robotic arm is determined to be F2. When the weight of the drone is equal to or greater than the second preset weight and less than the third preset weight, the basic thrust of the robotic arm is determined to be F3; When the weight of the drone is equal to or greater than the third preset weight and less than the fourth preset weight, the basic thrust of the robotic arm is determined to be F4.

[0007] Beneficial effects: By dividing the weight of the drone into different preset ranges and setting multiple levels of basic thrust accordingly, the robotic arm can be provided with graded and adapted basic forces according to the force requirements of drones of different weights. This avoids excessive or insufficient thrust caused by the mismatch between the thrust and the weight of the drone, making the support and adjustment of the drone by the robotic arm more stable and reliable, and further improving the stability and accuracy of the drone's return process.

[0008] In one optional implementation, when the position deviation is less than or equal to a second preset deviation, the actual thrust is 50% to 70% of the base thrust.

[0009] Beneficial effects: When the position deviation is less than or equal to the second preset deviation, the actual thrust is set to 50% to 70% of the base thrust. This allows for fine adjustments with a smaller thrust when the drone approaches the target position, avoiding overshoot, oscillation, or secondary deviation caused by excessive thrust. This makes the return process smoother and more precise, further improving the drone's docking stability.

[0010] In one optional implementation, when the position deviation is greater than a second preset deviation and less than or equal to a third preset deviation, the actual thrust is 70% to 90% of the base thrust.

[0011] Beneficial effects: When the position deviation is greater than the second preset deviation and less than or equal to the third preset deviation, the actual thrust is set to 70% to 90% of the base thrust. This can provide a moderate and stable adjustment thrust when the UAV has a moderate position deviation, taking into account both adjustment speed and motion smoothness. It avoids slow return to position due to insufficient thrust or attitude swaying due to excessive thrust, and further improves the accuracy and reliability of the UAV's return process.

[0012] In one optional implementation, when the position deviation is greater than a third preset deviation and less than or equal to a fourth preset deviation, the actual thrust is 90% to 100% of the base thrust.

[0013] Beneficial effects: When the position deviation is greater than the third preset deviation and less than or equal to the fourth preset deviation, the actual thrust is set to 90%~100% of the base thrust. This can provide sufficient adjustment thrust when the UAV position deviation is large, quickly reduce the deviation, and avoid the UAV attitude instability caused by full thrust impact, thus balancing homing efficiency and adjustment safety.

[0014] In one optional implementation, when the position deviation is greater than a fourth preset deviation and less than or equal to a fifth preset deviation, the actual thrust is the base thrust.

[0015] Beneficial effects: When the position deviation is greater than the fourth preset deviation and less than or equal to the fifth preset deviation, the actual thrust is directly set as the base thrust. This can provide sufficient adjustment driving force for the robotic arm when the deviation is in a large range, quickly and efficiently correcting the UAV to the target position, ensuring the return response speed and adjustment force, while maintaining the stability of the thrust output.

[0016] In one optional implementation, when the position deviation is greater than a fifth preset deviation, the actual thrust is divided into two stages. Initially, the actual thrust is 80% of the base thrust. After the position deviation is greater than a fourth preset deviation and less than or equal to a fifth preset deviation, the actual thrust is adjusted to the base thrust.

[0017] Beneficial effects: For situations where the position deviation is greater than the fifth preset deviation, the actual thrust is set in stages. In the initial stage, 80% of the base thrust is used for pre-adjustment. Once the deviation enters the fourth to fifth preset deviation range, the base thrust is switched. This allows for the use of gentle thrust to prevent the UAV from shaking violently when the deviation is large. Then, the thrust is gradually increased to achieve efficient correction. This ensures both rapid return to position efficiency and prevents attitude instability caused by sudden thrust changes, further improving the stability and accuracy of the UAV's return to position process.

[0018] In one optional implementation, moving the drone to the target location includes: The drone is moved to a position where the deviation is less than or equal to the first preset deviation.

[0019] Beneficial effects: By adjusting the drone to a position deviation less than or equal to the first preset deviation, the drone can accurately dock within the target location range, achieving high-precision homing and positioning. This effectively avoids docking failure or subsequent operational interference caused by excessive position deviation, further improving the accuracy and reliability of drone landing and recovery.

[0020] In one optional implementation, obtaining the weight of the drone includes: Obtain the model number of the drone and retrieve the weight information corresponding to that model, or obtain the signal from the weight sensor.

[0021] Beneficial effects: By retrieving the corresponding weight information based on the drone model or directly collecting weight sensor signals to obtain the drone's weight, the actual weight parameters of the drone can be obtained flexibly and accurately. This provides reliable data support for the precise determination of the robotic arm's thrust, ensuring that the thrust control matches the actual state of the drone and improving the accuracy and adaptability of the homing adjustment.

[0022] In one optional implementation, obtaining the landing location coordinates of the UAV includes: Obtain the three-dimensional coordinate data of the drone's fuselage outline.

[0023] Beneficial effects: By obtaining the three-dimensional coordinate data of the drone's fuselage outline to determine the landing position coordinates, the actual attitude and spatial position of the drone can be reflected more comprehensively and intuitively, avoiding the errors caused by single-point positioning, improving the accuracy and reliability of position detection, providing a stable and accurate position basis for subsequent robotic arm thrust calculation and precise correction, and further improving the drone's return accuracy.

[0024] Secondly, the present invention also provides a drone hangar, including a landing pad and a robotic arm disposed on the landing pad, the drone hangar being configured to perform the drone return method described in any of the above claims.

[0025] Beneficial effects: By setting up a robotic arm on the tarmac and implementing the aforementioned drone return method, the drone hangar can adaptively adjust the robotic arm thrust according to the drone's weight and real-time position deviation, achieving precise and stable return control of the drone. This effectively avoids problems such as drone deviation, slippage, or tipping during docking, significantly improving the stability, accuracy, and safety of automated drone recovery and docking, and meeting the needs of efficient and reliable automated parking management for multiple drone models.

[0026] In one alternative implementation, the robotic arm includes a first arm, a second arm, a third arm, and a fourth arm that are independent of each other, and the first arm, the second arm, the third arm, and the fourth arm are distributed around the helipad.

[0027] Beneficial effects: By using four independent robotic arms distributed around the helipad—the first, second, third, and fourth arms—instead of the traditional stepper motor and lead screw structure, the helipad is not worn during the drone's return to its position. The robotic arms are also more stable under force and will not deform due to different drone weights, resulting in greater durability. At the same time, it avoids the shortcomings of servo motor control, stepper motors, such as complexity, easy overload, lack of feedback, easy step loss, and poor positioning accuracy. The drone displacement control is achieved through the cooperation of multiple independent robotic arms, resulting in a more reliable overall structure, smoother adjustment, and higher positioning accuracy, which can adapt to the return requirements of drones of different weights. Attached Figure Description

[0028] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0029] Figure 1 This is a flowchart of a drone repositioning method according to an embodiment of the present invention; Figure 2 This is a perspective view of a drone hangar according to an embodiment of the present invention; Figure 3 This is a top view of a drone hangar according to an embodiment of the present invention, showing the robotic arm in an extended state; Figure 4 This is a top view of a drone hangar according to an embodiment of the present invention, showing the robotic arm in a hidden state; Figure 5 This is a schematic diagram of the connection structure of a drone homing control device according to an embodiment of the present invention.

[0030] Explanation of reference numerals in the attached figures: 1. Helipad; 2. Robotic arm. Detailed Implementation

[0031] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0032] The following is combined with Figures 1 to 5 The following describes embodiments of the present invention.

[0033] According to embodiments of the present invention, in one aspect, such as Figure 1 As shown, a method for homing a drone is provided, including: Obtain the landing coordinates of the drone; If the positional deviation between the landing location coordinates and the target location coordinates is greater than a first preset deviation, then the weight of the drone is obtained; Determine the basic thrust of robotic arm 2 based on the weight of the drone; The position deviation is divided into multiple intervals, and the actual thrust of the robotic arm 2 is determined based on the base thrust according to the interval where the position deviation is located. The robotic arm 2 applies a determined actual thrust to the drone, moving the drone to the target position.

[0034] By acquiring the landing position coordinates of the drone and further acquiring the drone's weight when the position deviation exceeds the limit, the basic thrust of robotic arm 2 is determined based on the weight. Then, the basic thrust is adaptively adjusted in combination with the position deviation range to obtain a suitable actual thrust to control the movement of robotic arm 2. After the drone lands, it can output a matching force based on its actual weight and real-time position deviation, effectively avoiding problems such as drone position deviation, attitude deviation, slippage, or even tipping caused by mismatch between the magnitude and orientation of the force. This significantly improves the drone's return accuracy and stability, reliably achieving precise adjustment and stable docking of the drone to the target position, and meeting the requirements of automated, high-precision landing and recovery.

[0035] In some embodiments, determining the basic thrust of the robotic arm 2 based on the weight of the drone includes: When the weight of the drone is less than the first preset weight, the basic thrust of the robotic arm 2 is determined to be F1; When the weight of the drone is equal to or greater than the first preset weight and less than the second preset weight, the basic thrust of the robotic arm 2 is determined to be F2. When the weight of the drone is equal to or greater than the second preset weight and less than the third preset weight, the basic thrust of the robotic arm 2 is determined to be F3. When the weight of the drone is equal to or greater than the third preset weight and less than the fourth preset weight, the basic thrust of the robotic arm 2 is determined to be F4.

[0036] By dividing the weight of the drone into different preset ranges and setting multiple levels of basic thrust accordingly, the robotic arm 2 can provide graded and adapted basic forces according to the force requirements of drones of different weights. This avoids excessive or insufficient thrust caused by the mismatch between the thrust and the weight of the drone, making the support and adjustment of the drone by the robotic arm 2 more stable and reliable, and further improving the stability and accuracy of the drone's return process.

[0037] In some embodiments, when the position deviation is less than or equal to a second preset deviation, the actual thrust is 50% to 70% of the base thrust.

[0038] When the position deviation is less than or equal to the second preset deviation, the actual thrust is set to 50% to 70% of the base thrust. This allows for fine adjustments with a smaller thrust when the drone approaches the target position, avoiding overshoot, oscillation, or secondary deviation caused by excessive thrust. This makes the return process smoother and more precise, further improving the drone's docking stability.

[0039] When the position deviation is greater than the second preset deviation and less than or equal to the third preset deviation, the actual thrust is 70% to 90% of the base thrust.

[0040] When the position deviation is greater than the second preset deviation and less than or equal to the third preset deviation, the actual thrust is set to 70% to 90% of the base thrust. This provides a moderate and stable adjustment thrust when the UAV has a moderate position deviation, taking into account both adjustment speed and motion smoothness. It avoids slow return to position due to insufficient thrust or attitude swaying due to excessive thrust, thereby further improving the accuracy and reliability of the UAV's return process.

[0041] When the position deviation is greater than the third preset deviation and less than or equal to the fourth preset deviation, the actual thrust is 90% to 100% of the base thrust.

[0042] When the position deviation is greater than the third preset deviation and less than or equal to the fourth preset deviation, the actual thrust is set to 90%~100% of the base thrust. This can provide sufficient adjustment thrust when the UAV position deviation is large, quickly reduce the deviation, and avoid the UAV attitude instability caused by full thrust impact, thus balancing return efficiency and adjustment safety.

[0043] When the position deviation is greater than the fourth preset deviation and less than or equal to the fifth preset deviation, the actual thrust is the base thrust.

[0044] When the position deviation is greater than the fourth preset deviation and less than or equal to the fifth preset deviation, the actual thrust is directly set as the base thrust. This provides sufficient adjustment driving force for the robotic arm 2 when the deviation is in a large range, quickly and efficiently correcting the UAV to the target position, ensuring the return response speed and adjustment force, while maintaining the stability of the thrust output.

[0045] When the position deviation is greater than the fifth preset deviation, the actual thrust is divided into two stages. Initially, the actual thrust is 80% of the base thrust. After the position deviation is greater than the fourth preset deviation and less than or equal to the fifth preset deviation, the actual thrust is adjusted to the base thrust.

[0046] For situations where the position deviation exceeds the fifth preset deviation, the actual thrust is set in stages. In the initial stage, 80% of the base thrust is used for pre-adjustment. Once the deviation enters the fourth to fifth preset deviation range, the base thrust is switched. This allows for the use of gentle thrust to prevent the UAV from shaking violently when the deviation is large, and then the thrust is gradually increased to achieve efficient correction. This ensures both rapid return to position efficiency and prevents attitude instability caused by sudden thrust changes, further improving the stability and accuracy of the UAV's return to position process.

[0047] It should be noted that moving the drone to the target location includes: Move the drone to a position deviation less than or equal to the first preset deviation.

[0048] By adjusting the drone's position deviation to be less than or equal to the first preset deviation, the drone can accurately dock within the target location range, achieving high-precision homing and positioning. This effectively avoids docking failure or subsequent operational interference caused by excessive position deviation, further improving the accuracy and reliability of drone landing and recovery.

[0049] In some embodiments, obtaining the weight of the drone includes: Obtain the drone model and retrieve the corresponding weight information, or obtain the signal from the weight sensor.

[0050] By retrieving the corresponding weight information based on the drone model or directly collecting weight sensor signals to obtain the drone's weight, the actual weight parameters of the drone can be obtained flexibly and accurately. This provides reliable data support for the precise determination of the thrust of the robotic arm 2, ensuring that the thrust control matches the actual state of the drone and improving the accuracy and adaptability of the homing adjustment.

[0051] The drone model can be obtained through either automatic identification or manual input. Automatic identification methods include: (1) QR code or barcode scanning and recognition Hardware configuration: Four sets of high-definition barcode scanning modules (with supplementary lighting function to adapt to low light or backlight environment) are embedded at the edge of hangar apron 1, and a mini barcode scanner is integrated at the end of robotic arm 2. Identification Process: Before the drone returns, it sends a landing warning to the hangar via the ground station or its own positioning system, and the hangar activates the scanning module in advance. After the drone lands, the scanning module on the helipad 1 scans the model QR code on the drone's fuselage from different angles (usually affixed to easily identifiable locations such as the landing gear and fuselage belly), and the end effector of the robotic arm 2 assists in scanning blind spots. The scanned data is transmitted to the CPU in real time via the RS485 interface. The CPU parses the model code in the QR code (such as "DJI-M300" or "AUTEL-EVO3") to complete the model identification.

[0052] (2) Infrared feature matching and database identification Hardware configuration: A 360° panoramic infrared imager is installed on the top of the hangar, with a resolution >1080P and a sampling frequency >30 frames / second; Identification Process: After the drone lands, the infrared imager quickly collects infrared feature data such as the drone's fuselage outline, landing gear spacing, and fuselage dimensions. The CPU compares the collected feature parameters with a pre-stored drone model-infrared feature database (the database contains information such as standard dimensions, outline features, and landing gear layout of mainstream drone models). The drone model is determined by a similarity algorithm (threshold 95%). If there is no matching model in the database, the system automatically marks it as "unknown model" and triggers a manual input prompt.

[0053] (3) Wireless communication protocol identification Supported protocols: Compatible with mainstream drone communication protocols such as DJISDK, Autel SkyLink, and MavLink, as well as custom proprietary protocols; Identification process: After the drone enters the hangar's communication range (50-meter radius), it automatically establishes a wireless connection with the hangar's control system and obtains basic information such as the drone model and firmware version through a protocol handshake process. This method is suitable for drones with intelligent communication capabilities, with an identification delay of 1 second.

[0054] The manual input method for obtaining the drone model is explained in detail below: Operating interface: The hangar is equipped with a touch screen (installed on the outside of the hangar), which supports manual input of model code or selection of common models from drop-down menu; it also supports remote input via PC and mobile APP, and data is synchronized to the controller via Ethernet / 4G module; Data storage: Manually entered model information is automatically associated with the current homing task. If the same model of drone is used repeatedly, the system automatically caches it locally, so that it can be selected directly next time without having to re-enter it.

[0055] It should be noted that if automatic recognition fails (e.g., QR code is damaged, infrared feature does not match, communication is interrupted), the system will issue an audible and visual alarm (flashing red light + buzzer prompt), and display "Recognition failed, please manually enter the model number" on the touch screen. If it still cannot match after manual input, the system will default to performing a return operation according to the parameters of a general drone. New models can be added later by updating the database in the background.

[0056] In some embodiments, obtaining the landing location coordinates of the drone includes: The three-dimensional coordinate data of the UAV's fuselage outline are obtained by scanning the helipad 1 with a lidar.

[0057] Determining the landing position coordinates by acquiring the three-dimensional coordinate data of the drone's fuselage outline can more comprehensively and intuitively reflect the drone's actual attitude and spatial position, avoid errors caused by single-point positioning, improve the accuracy and reliability of position detection, provide a stable and accurate position basis for subsequent thrust calculation and precise correction of the robotic arm, and further improve the drone's repositioning accuracy.

[0058] Two sets of LiDAR sensors (scanning range 0-360°, ranging accuracy: 1cm, scanning frequency 50Hz) are installed on the top of the hangar. Positioning markers (alternating black and white circular markers, 5cm in diameter, spaced 30cm apart) based on machine vision are laid on the surface of helipad 1. After the UAV lands, the LiDAR sensors quickly scan the helipad 1 area to obtain the three-dimensional coordinate data of the UAV's fuselage outline (establishing a Cartesian coordinate system with the center of helipad 1 as the origin, the X-axis along the length of helipad 1, the Y-axis along the width, and the Z-axis perpendicular to the plane of helipad 1). Simultaneously, the machine vision system identifies the relative position of the UAV landing gear and the positioning markers, corrects the positioning deviation of the LiDAR sensors, and finally outputs the precise coordinates (accuracy 2mm) of the four landing gear support points, which are then synchronized to the PLC.

[0059] The landing position coordinates are classified into the following five categories, based on the deviation between the actual landing position of the UAV and the target coordinates, with the center of landing pad 1 (0,0) as the target positioning coordinate:

[0060] The force control and azimuth control methods classified according to positional deviation are as follows: When making a precise landing (x ≤ ±5 cm and y ≤ ±5 cm), the micro-thrust calibration mode is adopted. Each robotic arm 2 applies 50% - 70% of the basic thrust to avoid excessive thrust causing the drone to shake. The robotic arm 2 extends from the internal slots at the four corners of the landing pad 1 and aligns with the stress points at the four corners of the drone's landing gear (preset rubber cushions to increase friction). The thrust direction is along the straight line from the "landing gear stress point - the center of the landing pad 1". The four robotic arms 2 exert force simultaneously at a translation speed of 1 cm / s until the center point of the drone's landing gear coincides with the center (0, 0) of the landing pad 1, and stops when the deviation is 1 mm.

[0061] When there is a slight deviation (5 cm < x ≤ 15 cm or 5 cm < y ≤ 15 cm), 70% - 90% of the basic thrust is applied. The greater the deviation, the closer the thrust is to the upper limit (for example, when x = 15 cm, the thrust is 90% of the basic thrust). The primary and secondary thrust cooperation mode is adopted. The two robotic arms 2 in the deviation direction are the main thrust arms, and the two robotic arms 2 on the other side are the auxiliary calibration arms. For example, if the drone is biased towards the positive X-axis (x = 12 cm, y = 4 cm), the two robotic arms 2 in the negative X-axis direction apply the main thrust (90% of the basic thrust), and the two robotic arms 2 in the positive X-axis direction apply the auxiliary thrust (70% of the basic thrust). The robotic arms 2 in the Y-axis direction are adjusted synchronously to ensure that the return trajectory is a straight line and avoid lateral deviation.

[0062] When there is a moderate deviation (15 cm < x ≤ 30 cm or 15 cm < y ≤ 30 cm), 90% - 100% of the basic thrust is applied, and it is dynamically adjusted according to the weight of the drone (the upper limit is taken for heavy drones, and the upper-middle limit is taken for light drones). It is executed in two steps. The first step is coarse adjustment to the center: The two robotic arms 2 in the deviation direction exert force to translate the drone to a range with a deviation of 10 cm at a translation speed of 2 cm / s. The second step is fine adjustment and calibration: Switch to the force control mode of precise landing, and the four robotic arms 2 exert force synchronously to fine-tune to the target coordinates. If the drone has a slight tilt (Z-axis deviation of 3 cm), the thrust direction of the robotic arm 2 is slightly offset in the opposite direction of the tilt (offset angle = body tilt angle) to ensure that the body gradually levels during the return process.

[0063] When there is a severe deviation (30 cm < x ≤ 50 cm or 30 cm < y ≤ 50 cm), 100% of the basic thrust is applied. Super-heavy drones can superimpose a compensation thrust of 5 - 8 N to ensure sufficient power. The segmented translation plus attitude calibration mode is adopted. The unilateral robotic arm 2 first exerts force to pull the drone back to the central area of the landing pad 1 (deviation of 20 cm) at a translation speed of 3 cm / s to avoid the drone tipping over due to being too fast. Then, adjust the thrust directions of the four robotic arms 2 to correct the body attitude (horizontal + facing the center). Then, perform fine adjustment to return to the target coordinates. If abnormal resistance (> 1.5 times the basic thrust) is detected by the pressure sensor during the return process, the system immediately stops the thrust output and issues an alarm to avoid damage to the robotic arm 2 or the drone.

[0064] When the extreme deviation occurs (x>50cm or y>50cm), 80% of the initial base thrust is applied. Once the drone enters the severe deviation range, the force control standard for severe deviation is switched. The robotic arm 2 extends to its maximum stroke (80cm), grabs the drone's landing gear, and slowly pulls it towards the center of the apron 1 (pull speed 2cm / s). During this process, coordinate changes are monitored in real time, and the thrust direction is calibrated every 10cm of translation to ensure the trajectory remains unchanged. If the drone exceeds the edge of apron 1 by more than 10cm, the system first applies a pullback thrust through the outer robotic arm 2 to pull the drone back to the apron 1 area before executing subsequent return-to-position steps. Direct lateral translation is prohibited.

[0065] During the application of force by robotic arm 2, the thrust is monitored in real time, and feedback and dynamic adjustments are made. Each robotic arm 2 has a built-in force sensor (accuracy +0.1N) that feeds back the actual thrust value to the PLC in real time. The PLC compares the deviation between the actual thrust and the target thrust (allowable deviation 0.5N) and dynamically adjusts the output torque of the servo motor through a PID algorithm. If the thrust deviation is >1N for 3 consecutive seconds, or if the drone experiences abnormal displacement (such as a sudden change in Z-axis height >5cm), the system immediately cuts off the power to robotic arm 2, stops the return operation, issues an audible and visual alarm, and awaits manual investigation.

[0066] After synchronously acquiring the location coordinates and weight information of the UAV landing pad 1, the PLC accurately returned to its original position, as detailed below: First, establish a unified coordinate system with the geometric center of landing pad 1 as the origin (0,0). Establish a two-dimensional Cartesian coordinate system, with the X-axis along the length of landing pad 1 and the Y-axis along its width. The coordinate unit is millimeters (mm). The PLC receives the original coordinates of the four support points of the UAV landing gear from laser scanning or visual positioning transmission, denoted as P1(x1,y1), P2(x2,y2), P3(x3,y3), and P4(x4,y4). Calculate the coordinates of the geometric center point of the landing gear, P0(x0,y0), using the following formula: x0 = (x1 + x2 + x3 + x4) / 4; y0 = (y1 + y2 + y3 + y4) / 4; Deviation quantification: Calculate the centering deviation Δx=x0-0, Δy=y0-0 of the center point P0 relative to the origin (0,0); at the same time, calculate the offset of the four support points relative to the center point P0 (Δx1=x1-x0, Δy1=y1-y0) to ensure that the attitude of the UAV remains unchanged during the centering process.

[0067] The PLC receives the total weight W (or manually input weight) collected by the pressure sensor, removes outliers (e.g., if the deviation of a single collected value from the average value is >10%, it re-collects the weight), and finally outputs the valid weight value Wo. Wo is divided into 4 levels and matched with a preset force control parameter library (the parameter library contains basic thrust, acceleration threshold, and friction coefficient compensation values ​​corresponding to different weights). The specific classifications are as follows:

[0068] The PLC determines the centering difficulty level based on the absolute values ​​of the centering deviations (Δx, Δy) and the weight level, and matches differentiated motion control strategies accordingly. Based on the deviation model and weight level, the PLC calculates the target displacement, thrust magnitude, and direction of each robotic arm 2 through dynamic equations and inverse kinematics. The core calculation steps are as follows: 1. Target displacement calculation (end-effector trajectory of robotic arm 2) The four robotic arms 2 move synchronously, ultimately moving the four support points of the drone's landing gear to the target coordinates P1'(x1',y1'), P2'(x2',y2'), P3'(x3',y3'), and P4'(x4',y4') (with the target center point as the origin and the relative positions of the support points remaining unchanged); Displacement calculation for a single robotic arm 2: Taking the support point P1 of robotic arm 2 as an example, the formula for the target displacement (Δx1,Δy1) is: Δx1 =(0-x0)+ Δx1 =-x0 +(x1 -x0)=x1-2x0; Δy1=(0-y0)+Δy1=-y0 +(y1-y0)=y1-2y0; Similarly, calculate the target displacements (Δx2, Δy2), (Δx3, Δy3), and (Δx4, Δy4) of robotic arms two, three, and four; adopt an S-shaped acceleration and deceleration trajectory to avoid start-stop impact, and decompose the displacement into an acceleration segment, a constant speed segment, and a deceleration segment. The constant speed segment speed v is set according to the weight class (lightweight v=20mm / s, heavy-duty v=5mm / s).

[0069] 2. Target thrust calculation (dynamic force control based on weight and deviation) The basic thrust of each robotic arm 2 is F0 = W0 × k (where k is the thrust coefficient for the corresponding weight class), ensuring that the thrust is sufficient to overcome the inertia of the drone and the friction of the contact surface; the compensation thrust ΔF is added according to the centering difficulty level, with a larger compensation ratio for higher difficulty levels. F = F0 × (1 + ΔF) 比例 ); Among them, the first level of difficulty ΔF 比例 =0.1, Level 2 difficulty ΔF 比例 =0.2, Level 3 difficulty ΔF 比例=0.3, Level 4 difficulty ΔF 比例 =0.5; The thrust direction is along the straight line from the current position of the fulcrum to the target position. The PLC calculates the angle θ between the thrust and the X-axis using trigonometric functions: θ = arctan(Δyi / Δxi) (where i = 1~4, corresponding to the four robotic arms 2 respectively), ensuring that the thrust direction of the four robotic arms 2 all points to the target position of their respective fulcrums, forming a coordinated centering force.

[0070] 3. Attitude correction and compensation (adapting to tilt / eccentricity situations) If the laser scan detects that the drone's fuselage is tilted (tilt angle a ≤ 15°), the PLC calculates the correction torque M = W0 × g × d (d is the center of gravity offset) based on the weight distribution, and adjusts the thrust difference (ΔF) of the diagonal robotic arm 2 accordingly. 修正 =M / L, where L is the distance between fulcrums), and the fuselage attitude is corrected synchronously during the centering process; If the drone's center of gravity is off-center (e.g., due to uneven load), the pressure sensor will report the pressure difference at each support point, and the PLC will distribute the thrust according to the pressure ratio (the support point with higher pressure will receive slightly more thrust from robotic arm 2) to prevent the drone from tipping over during the centering process.

[0071] Command allocation: Robotic arm 2 issues commands precisely.

[0072] 1. Instruction encoding and format The PLC encapsulates the calculated displacement, thrust, direction angle, motion speed, and start / stop timing into standardized instruction frames.

[0073] 2. Distribution and Execution Logic The PLC sends commands synchronously to the servo drives of the four robotic arms 2 via the CAN bus (transmission delay 1ms) to ensure that all robotic arms 2 start at the same time t. After receiving the command, the servo driver amplifies the torque through the reducer and drives the two joints of the robotic arm to move along the planned trajectory. At the same time, it feeds back the actual displacement and thrust data to the PLC in real time.

[0074] The PLC employs a closed-loop control logic of command issuance, data feedback, and parameter correction to dynamically adjust the movements of robotic arm 2 and avoid error accumulation. 1. Feedback Data Collection Each robotic arm 2 has a built-in displacement sensor (accuracy ±0.1mm) and force sensor (accuracy ±0.1N), transmitting the actual displacement (xi) to the PLC every 10ms. 实 yi 实 ) and actual thrust Fi 实 ; 2. Error Calculation PLC calculates the deviation (Δx) between the actual value and the target value. 误差=xi 实 -Δxi,ΔF 误差 =Fi 实 -Fi); 3. Dynamic correction If Δx 误差 >±0.5mm or ΔF 误差 With a torque of >±0.3N, the PLC adjusts the output torque of the servo motor through the PID algorithm, correcting the displacement and thrust until the deviation of the UAV's center point is ≤±1mm, and the return to position is completed.

[0075] According to an embodiment of the present invention, in another aspect, a drone hangar is also provided, such as... Figures 2-4 As shown, the drone hangar includes a helipad 1 and a robotic arm 2 installed on the helipad 1. The drone hangar is configured to perform the above-described drone return method.

[0076] By installing a robotic arm 2 on the helipad 1 and executing the aforementioned drone return method, the drone hangar can adaptively adjust the thrust of the robotic arm 2 according to the drone's weight and real-time position deviation, achieving precise and stable return control of the drone. This effectively avoids problems such as drone deviation, slippage, or tipping during docking, significantly improving the stability, accuracy, and safety of automated drone recovery and docking, and meeting the needs of efficient and reliable automated parking management for multiple drone models.

[0077] In some embodiments, the robotic arm 2 includes a first arm, a second arm, a third arm, and a fourth arm that are independent of each other, and the first arm, the second arm, the third arm, and the fourth arm are distributed around the helipad 1.

[0078] The system employs four independent robotic arms 2, distributed around the helipad 1 (first, second, third, and fourth arms), replacing the traditional stepper motor and lead screw structure. This design prevents wear and tear on the helipad 1 during the drone's return to its position, and the robotic arms 2 maintain stable force distribution, preventing deformation due to varying drone weights and enhancing durability. Furthermore, it avoids the drawbacks of servo motor control, stepper motors (such as overload, lack of feedback, step loss, and poor positioning accuracy). By coordinating multiple independent robotic arms 2, the system achieves drone displacement control, resulting in a more reliable overall structure, smoother adjustments, and higher positioning accuracy, thus adapting to the return requirements of drones of varying weights.

[0079] In some embodiments, during operation, four robotic arms 2 grasp the four corners of the drone's landing gear and place the drone in a pre-set designated area. When not in operation, the robotic arms 2 are hidden in the hangar, and are not visible from above the helipad 1, thus having no impact on the drone's takeoff and landing. The drone does not come into contact with the surface of the helipad 1 during movement, resulting in no wear and tear and extending its service life.

[0080] The drone hangar assists drones in accurately returning to their designated positions. The entire system consists of a control system, a drive system, and an actuator.

[0081] Preferably, the control system uses a PLC to issue commands, receives data from the pressure sensor to make corresponding judgments, and calculates and compensates for the difference between the target force and torque and the actual force in real time.

[0082] Preferably, the drive device is a servo motor, which amplifies the torque through a reducer to drive the joint rotation, thereby achieving translation of the UAV in various directions.

[0083] like Figure 5 As shown, during drone operations, the hangar control system obtains the drone's model and landing coordinates. The method of acquisition is not limited; it can be obtained manually or through laser scanning. The CPU receives the drone's model and location information from the input unit. The CPU retrieves the drone's mass information from memory and combines it with the location coordinates for calculation and analysis. The output unit then controls the magnitude and direction of the force applied by each robotic arm 2. Similarly, after obtaining the drone model, the CPU retrieves the corresponding QR code pattern from memory and projects it onto the landing pad 1 via the I / O interface to facilitate drone landing.

[0084] After the drone returns to base and lands, the PLC will determine whether to obtain the drone model and location coordinates.

[0085] If not, the hangar will perform the identification again; If so, the two pieces of information are passed to the PLC for processing.

[0086] For drones, the hangar can obtain the drone's accurate mass through infrared scanning and transmit it to the PLC. This information can also be manually entered; there are no restrictions on this. After landing on helipad 1, the drone's location coordinates are also transmitted to the PLC.

[0087] After receiving two pieces of information, the PLC processes them and issues instructions to the four robotic arms 2, specifying their respective extension positions and the direction and force to be applied, thereby enabling precise repositioning.

[0088] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations all fall within the scope defined by this application.

Claims

1. A method for returning a drone to its designated location, characterized in that, include: Obtain the landing location coordinates of the drone; If the positional deviation between the landing position coordinates and the target position coordinates is greater than a first preset deviation, then the weight of the drone is obtained; Determine the basic thrust of the robotic arm based on the weight of the drone; The position deviation is divided into multiple intervals, and the actual thrust of the robotic arm is determined based on the base thrust according to the interval where the position deviation is located. The robotic arm is controlled to apply a determined actual thrust to the drone, moving the drone to the target position.

2. The drone repositioning method according to claim 1, characterized in that, Determining the basic thrust of the robotic arm based on the weight of the drone includes: When the weight of the drone is less than the first preset weight, the basic thrust of the robotic arm is determined to be F1; When the weight of the drone is equal to or greater than the first preset weight and less than the second preset weight, the basic thrust of the robotic arm is determined to be F2. When the weight of the drone is equal to or greater than the second preset weight and less than the third preset weight, the basic thrust of the robotic arm is determined to be F3; When the weight of the drone is equal to or greater than the third preset weight and less than the fourth preset weight, the basic thrust of the robotic arm is determined to be F4.

3. The drone repositioning method according to claim 1, characterized in that, When the position deviation is less than or equal to the second preset deviation, the actual thrust is 50% to 70% of the base thrust.

4. The drone repositioning method according to claim 1, characterized in that, When the position deviation is greater than the second preset deviation and less than or equal to the third preset deviation, the actual thrust is 70% to 90% of the base thrust.

5. The drone repositioning method according to claim 1, characterized in that, When the position deviation is greater than the third preset deviation and less than or equal to the fourth preset deviation, the actual thrust is 90% to 100% of the basic thrust.

6. The drone repositioning method according to claim 1, characterized in that, When the position deviation is greater than the fourth preset deviation and less than or equal to the fifth preset deviation, the actual thrust is the base thrust.

7. The drone repositioning method according to claim 1, characterized in that, When the position deviation is greater than the fifth preset deviation, the actual thrust is divided into two stages. Initially, the actual thrust is 80% of the base thrust. After the position deviation is greater than the fourth preset deviation and less than or equal to the fifth preset deviation, the actual thrust is adjusted to the base thrust.

8. The drone repositioning method according to claim 1, characterized in that, Moving the drone to the target location includes: The drone is moved to a position where the deviation is less than or equal to the first preset deviation.

9. The method for returning a drone to its original position according to any one of claims 1 to 8, characterized in that, The process of obtaining the weight of the drone includes: Obtain the model number of the drone and retrieve the weight information corresponding to that model, or obtain the signal from the weight sensor.

10. The drone repositioning method according to claim 1, characterized in that, Obtaining the landing location coordinates of the drone includes: Obtain the three-dimensional coordinate data of the drone's fuselage outline.

11. A drone hangar, characterized in that, The drone hangar includes a helipad (1) and a robotic arm (2) disposed on the helipad (1), and is configured to perform the drone return method according to any one of claims 1 to 10.

12. The drone hangar according to claim 11, characterized in that, The robotic arm (2) includes a first arm, a second arm, a third arm and a fourth arm that are independent of each other, and the first arm, the second arm, the third arm and the fourth arm are distributed around the helipad (1).

Images