A system for drone landing and charging in dynamic scenarios

Abstract

This invention discloses a system for landing and charging unmanned aerial vehicles (UAVs) in dynamic scenarios, relating to the field of unmanned system integration technology. It includes: a mobile main body, an attitude compensation platform, a landing platform, and a wireless charging module. The attitude compensation platform is rigidly connected to the landing platform to compensate for the attitude of the mobile main body. The landing platform provides a landing area for the UAV. A positioning device is installed on the landing platform to guide the UAV to locate the mobile main body. The transmitter of the wireless charging module is embedded in the landing platform, and the receiver is installed on the bottom of the UAV. When the UAV lands on the landing platform, the transmitter and receiver wirelessly connect and charge the UAV. When the UAV completes charging, charging ends and the UAV is unlocked. Using this invention can improve the success rate of UAV recovery in dynamic environments, achieve high-precision autonomous recovery in GPS-denied environments, and realize a fully automatic "landing and immediate charging" closed-loop system.

Description

Technical Field

[0001] This invention relates to the field of unmanned system integration technology, and in particular to a system for landing and charging unmanned aerial vehicles (UAVs) in dynamic scenarios. Background Technology

[0002] Currently, in high-risk operational scenarios such as disaster search and rescue, underground space exploration, and battlefield reconnaissance, heterogeneous collaboration between rotary-wing UAVs and quadruped robots is increasingly becoming a research hotspot in unmanned systems. However, in actual operations, rotary-wing UAVs are limited by battery capacity and have to frequently interrupt their missions to return to base for charging. This greatly limits their effective operating radius, especially in complex space exploration missions without GPS and with strong shielding, where data interaction between UAVs and ground stations is highly dependent on the communication relay node carried by the quadruped robot. This deeply coupled collaborative relationship can easily trigger a "chain reaction" of mission interruptions: once the UAV's energy is exhausted, the quadruped robot, which serves as the communication support, also has to abandon its current exploration progress and return to base to maintain the link.

[0003] To address systemic failures caused by endurance issues, Ropero et al. first proposed in 2019 that unmanned vehicles could serve as mobile charging stations to provide energy for drones when unmanned systems explore planetary surfaces. Unfortunately, this was only tested in simulations. Moore et al. (2023) proposed a "bag-type" recovery architecture, which initially achieved the recovery and recharging of drones on quadruped robots. However, its guidance mechanism has extremely high requirements for recovery attitude, resulting in a high landing failure rate.

[0004] Furthermore, the maximum tilt angle of back pitch and roll disturbances during quadruped robot movement can reach over 15°. Combined with the undulating terrain in field operations, a flat landing reference cannot be provided. Passive vibration damping is ineffective against low-frequency, large-amplitude tilts. Traditional serial three-axis gimbals have poor rigidity and a high center of gravity, severely affecting the quadruped robot's motion stability. Traditional USB wired charging is prone to loosening under continuous vibration in quadruped robot environments, and exposed cables interfere with the Stewart platform's servo motor movement. Manual plugging and unplugging is impractical in dynamic movement scenarios, making automated charging impossible. GPS completely fails indoors, underground, and in obstructed environments. Traditional visual landing markers (ArUco, etc.) are severely affected by lighting and occlusion. The continuous movement of the robot's back platform requires the positioning system to track the moving target in real time and perform calculations at the edge. Currently, there is no solution that simultaneously addresses the three coupled problems of platform tilt, contactless charging, and GPS refusal to locate in the dynamic mobile platform scenario of a continuously moving quadruped robot. Summary of the Invention

[0005] In order to solve the problems existing in the prior art, the present invention provides the following technical solution.

[0006] This invention provides a system for landing and charging a drone in a dynamic scenario, comprising: a mobile main body and an attitude compensation platform, a landing platform, and a wireless charging module located on the mobile main body. The attitude compensation platform is rigidly connected to the landing platform and is used to compensate for the attitude of the mobile main body, keeping the landing platform horizontal. The landing platform provides a landing area for the drone. A positioning device is provided on the landing platform to guide the drone to locate the mobile main body. The transmitter of the wireless charging module is embedded in the landing platform, and the receiver of the wireless charging module is installed on the bottom of the drone. When the drone lands on the landing platform, the transmitter and receiver are wirelessly connected and the drone is charged. When the drone finishes charging, the charging ends and the drone is unlocked.

[0007] Preferably, the positioning device guides the UAV to locate the moving subject using the following method: Using the current pose of the moving body as the origin, determine the local body coordinate system and the fixed offset of the landing platform in the local body coordinate system; Based on the real-time attitude of the moving subject, the corresponding rotation matrix is ​​determined; The coordinates of the landing platform in the global coordinate system are determined based on the real-time attitude, rotation matrix, and fixed offset.

[0008] Preferably, the coordinates of the landing platform in the global coordinate system are determined according to the following formula: in,( x world , y world , z world ( ) represents the coordinates of the landing platform in the global coordinate system; x c , y c , z c R(q) represents the coordinates of the moving subject in the global coordinate system; R(q) is the rotation matrix; q is the real-time attitude; dx , dy , dz ) represents a fixed offset.

[0009] Preferably, the system further includes a controller configured to execute the following program: In response to the drone's landing request, the system verifies the pose data of the mobile entity and the horizontal state of the landing platform. If the verification passes, the system acquires the positions of the drone and the mobile entity, calculates the horizontal deviation, generates a tracking control command, and drives the drone to follow the horizontally moving landing target point. If the horizontal deviation is less than a set threshold, the system drives the drone to initiate a vertical descent procedure and corrects the landing altitude target value in real time. If the distance between the drone and the landing platform reaches the determined landing altitude target value, a landing command is sent to make the drone contact the landing platform. Based on the contact information between the drone and the landing platform, the system drives the drone to automatically lock its motors and automatically activates the wireless charging module.

[0010] Preferably, the system further includes a first connecting platform located between the mobile body and the attitude compensation platform, which is used to achieve geometric alignment, stiffness transition and vibration isolation between the mounting interface of the mobile body and the base of the attitude compensation platform.

[0011] Preferably, the system further includes a second connection platform located between the attitude compensation platform and the landing platform, for receiving the output of the attitude compensation platform and expanding the area of ​​the UAV landing zone.

[0012] Preferably, the compensation platform is a six-degree-of-freedom parallel self-stabilizing platform, used to compensate for the motion of the moving body and the pitch and roll disturbances caused by the undulations of the ground in real time, so that the landing platform always maintains a dynamic horizontal state.

[0013] Preferably, the transmitter of the wireless charging module uses an independent power supply, which is isolated from the electrical system of the mobile body.

[0014] Preferably, the moving body is a quadruped robot.

[0015] Preferably, the positioning device uses one of the following: Lighthouse positioning extension board, UWB ultra-wideband, or indoor RTK.

[0016] Compared with the prior art, the present invention has the following beneficial effects: 1. Significantly improves the success rate of drone recovery in dynamic environments: The six-DOF Stewart platform actively compensates for gait disturbances and ground undulations of the Unitree Go2, keeping the platform tilt angle within approximately ±5°, which greatly improves the success rate of autonomous landing of Crazyflie 2.1 on the robot's back while it is moving.

[0017] 2. Achieve a fully automatic recharge closed loop: Based on the Qi standard wireless charging integration solution, eliminate the risk of connector loosening. The drone automatically starts charging after landing, and the robot dog can continue to move during charging, supporting multiple rounds of uninterrupted relay operations.

[0018] 3. Achieve high-precision autonomous retrieval in GPS-denied environments: Based on the pure laser retrieval algorithm using Lighthouse laser positioning (accuracy ±5mm, update frequency ≥100 Hz), it does not rely on GPS or visual recognition and can still reliably guide Crazyflie 2.1 to autonomous descent in GPS-deficient scenarios such as indoors, underground, and strong obstruction.

[0019] 4. High system integration and good reliability: The multi-layer structure design separates the mechanical and electrical interference of each functional module, and the distributed control architecture reuses the computing power of the robot body; the three-system (positioning-compensation-charging) collaborative closed-loop design makes the system robust in complex field environments.

[0020] 5. Lightweight, low cost, and easy integration: The total weight of the entire platform is approximately 900 g. It adopts a low-cost hardware combination of acrylic sheet + digital servo motor + ESP32, and can be applied to other models of quadruped robots after the adapter layer is replaced. Attached Figure Description

[0021] Figure 1 This is a schematic diagram of the system for drone landing and charging in dynamic scenarios as described in this invention; Figure 2 This is a schematic diagram of the assembly structure of the second connecting platform and the landing platform of the present invention; wherein, A is a perspective view, B is a top view, C is a left view, and D is a front view; The meanings of the symbols in the diagram are as follows: 1. Moving main body; 2. Connecting fastener; 3. First connecting platform; 4. Attitude compensation platform; 5. Second connecting platform; 6. Landing platform; 7. Circular groove; 8. Limiting device; 9. Lighthouse positioning extension plate mounting position; 10. Wiring groove and outlet. Detailed Implementation

[0022] To better understand the above technical solutions, the following will provide a detailed explanation of the technical solutions in conjunction with the accompanying drawings and specific implementation methods.

[0023] The problems that this invention aims to solve include: How to provide a dynamically level landing platform with active compensation for a drone during the dynamic movement of a quadruped robot, eliminating multi-axis disturbances caused by gait rhythm; how to achieve automatic charging and recharging of the drone after landing without physical contact, eliminating the risk of connector loosening and supporting the robot to continue moving during charging; how to use a positioning system to achieve high-precision real-time positioning and autonomous landing guidance of the drone relative to the back platform of the mobile robot in complex environments with GPS rejection and strong obstruction, without relying on visual recognition.

[0024] The technical content of this invention includes: (1) using the Stewart platform as the carrier layer of the UAV take-off and landing platform and rigidly integrating it with the wireless charging transmitter. In the prior art, the Stewart platform is used for camera or instrument stabilization, and there is no public record of its use for UAV dynamic landing compensation; (2) a dynamic target point algorithm for the mobile platform based on the quaternion transformation of the body coordinate system. This algorithm is designed specifically for the scenario of "the landing target itself is continuously moving", which is fundamentally different from the existing fixed platform landing guidance method in terms of problem definition; (3) the state machine sequence of landing contact signal → automatic charging start → flight control unlocking when fully charged. This sequence couples the mechanical contact event, the wireless charging state and the flight control unlocking logic, which is a system-level integrated innovation and can be obviously obtained by non-technical personnel based on the existing technology. The relevant content of the mechanical structure and control algorithm of the Stewart platform itself can be found in the prior patent (application number: 2026103240279, title: a six-degree-of-freedom parallel self-stabilizing platform and control system for quadruped robots).

[0025] This invention sets up an attitude compensation platform between the mobile body and the landing platform. The attitude compensation platform provides a dynamic level reference for the landing platform, improving the success rate and robustness of UAV landing and recovery on the landing platform. At the same time, a wireless charging module is integrated into the landing platform to achieve "recharging upon landing" without physical plugging and unplugging, thereby expanding the radius of collaborative operation tasks.

[0026] like Figure 1 As shown in the figure, this embodiment of the invention provides a system for landing and charging a drone in a dynamic scene, including: a mobile body 1 and an attitude compensation platform 4, a landing platform 6, and a wireless charging module (not shown) located on the mobile body 1. The attitude compensation platform 4 is rigidly connected to the landing platform 6. The attitude compensation platform 4 is used to compensate the attitude of the mobile body 1 so that the landing platform 6 remains in a horizontal state. The landing platform 6 is used to provide a landing area for the drone. A positioning device is provided on the landing platform 6 to guide the drone to locate the mobile body 1. The transmitter of the wireless charging module is embedded in the landing platform 6, and the receiver of the wireless charging module is installed on the bottom of the drone. When the drone lands on the landing platform 6, the transmitter and receiver are wirelessly connected and the drone is charged. When the drone finishes charging, the charging ends and the drone is unlocked.

[0027] The mobile platform (specifically a quadruped robot) can be based on the Unitree Go2 EDU quadruped robot, using its standard mounting slot on its back to mount the entire platform system. The quadruped robot's structure includes: a main body; a quadruped drive mechanism; an onboard power module; an onboard computing platform; a back mounting slot; and a communication module. The quadruped robot base provides a mobile takeoff, landing, and recharging platform for the drone. This structure can move stably in complex unstructured terrain, possessing strong obstacle-crossing and load-bearing capabilities. Simultaneously, this mobile base is also used for: ground operations; mounting radar and depth cameras; providing edge computing capabilities; serving as a drone recovery and transportation platform; and providing a mobile reference coordinate system for cooperative navigation.

[0028] The attitude compensation platform (a six-DOF Stewart parallel attitude compensation platform, for details please refer to the patent, application number: 2026103240279, title: A six-DOF parallel self-stabilizing platform and control system for quadruped robots) can compensate for the gait disturbance of quadruped robots in real time through active inverse kinematics calculation, and provide a dynamic horizontal reference for the upper landing platform. The drone landing platform (docking layer) can provide a take-off and landing contact surface with anti-slip limiting structure (limiting groove spacing is designed according to Crazyflie2.1 diagonal wheelbase of 92 mm) and Lighthouse positioning extension plate mounting position.

[0029] In one embodiment of the present invention, the structure of the drone landing platform may include: a take-off and landing contact surface; a guide edge structure; an anti-slip structure; a wireless charging transmitter coil mounting area; and a safety limiting structure. The drone landing platform provides a stable autonomous landing area for rotary-wing drones. This structure enables: autonomous landing guidance for the drone; take-off and landing buffering; landing position restriction; wireless charging transmitter integration; and drone parking fixation. By setting the anti-slip and limiting structures, the parking stability of the drone on the dynamic platform can be improved.

[0030] In one embodiment of the present invention, the assembly structure of the second connecting platform and the landing platform can be as follows: Figure 2 As shown. Among them, the circular groove 7 is used to embed the transmitter of the wireless charging module; the limiting device 8 is a four-corner or ring structure, used to constrain the landing position of the Crazyflie 2.1 drone and prevent the drone from slipping during landing; the Lighthouse positioning extension board mounting position 9 is located on the side extension plate of the main body, with a wide field of view to facilitate the reception of base station laser signals; the wiring slot and outlet 10 are used to provide a passage for the cable that independently powers the transmitter of the wireless charging module, so as to electrically isolate it from the Stewart control system.

[0031] In one embodiment of the present invention, the transmitter (Qi transmitting coil and matching circuit) of the wireless charging module can be embedded in the second connection platform and landing platform, rigidly connected to the Stewart platform moving platform, and moves synchronously with the attitude compensation of the moving platform to ensure that the transmitter and receiver always remain parallel and aligned under dynamic conditions. The operating frequency is 100 kHz, the maximum output power is 5 W, and it is compatible with commercially available Qi standard chargers. The receiver of the wireless charging module can refer to existing open-source solutions (Qi WPC v1.2 receiver circuit based on TIBQ51013B, referring to the Qi 1.2 Deck schematic diagram published by Bitcraze AB) and be integrated as a known prior art component of this system. The core components of the wireless charging module receiver consist of BQ51013B, DS28E05 single-bus authentication chip, and charging receiving coil, and are installed on the bottom of the Crazyflie 2.1 drone. The BQ51013B integrates full-bridge synchronous rectification, LDO regulation (5 V / 1 A output), foreign object detection (FOD), overcurrent / overtemperature protection, and charging status feedback (CHG pin).

[0032] The complete circuit workflow of the wireless charging module is as follows: the transmitter outputs a 100 kHz alternating magnetic field, which is induced by the coil at the receiver to generate an alternating voltage. After resonant frequency selection by the front-end matching capacitor, the voltage is input to the BQ51013B. The internal full-bridge synchronous rectification of the chip converts the AC power to DC power, which is then regulated to 5V by an LDO (low dropout linear regulator) to charge the Crazyflie 2.1 battery. Overcurrent, overtemperature, and foreign object abnormalities are monitored in real time throughout the charging process. The CHG pin feeds back the charging status to the flight controller, which completes device authentication by reading the DS28E05 via a single bus.

[0033] The wireless charging module provided by this invention has a Qi transmitting coil embedded in a dynamic landing platform, rigidly connected to the Stewart dynamic platform, with attitude compensation and coil alignment moving synchronously. After Crazyflie 2.1 completes autonomous landing and the flight controller detects a contact signal, it automatically locks the motors, and the wireless charging transmitter automatically starts without manual intervention. The CHG pin of the BQ51013B provides real-time feedback on the charging status to the flight controller, and the flight controller automatically enters the unlocking preparation state after full charge. The wireless charging transmitter is independently powered and electrically isolated from the Stewart control system, allowing the robot dog to continue moving during charging.

[0034] In a preferred embodiment of the present invention, the system further includes a first connecting platform 3, which is located between the mobile body 1 and the attitude compensation platform 4. The first connecting platform 3 is used to achieve geometric alignment, stiffness transition, and vibration isolation between the mounting interface of the mobile body 1 and the base of the attitude compensation platform 4. In one example, the first connecting platform is used to realize the mechanical connection and structural transition between the quadruped robot and the six-degree-of-freedom Stewart platform. The main functions of this structure include: aligning the mounting groove on the robot's back with the opening of the parallel platform; providing overall structural stiffness support; distributing the load of the parallel platform; absorbing robot gait vibrations; and reserving power and communication line channels. By setting up a vibration isolation structure, the impact of high-frequency vibrations generated during quadruped robot walking on the stability of UAV recovery can be reduced.

[0035] In another example, a connecting fastener 2 is also provided between the first connecting platform and the back of the quadruped robot for mounting the first connecting platform on the back of the quadruped robot.

[0036] In another preferred embodiment of the present invention, the system further includes a second connection platform 5, which is located between the attitude compensation platform 4 and the landing platform 6, and is used to receive the output of the attitude compensation platform 4 to expand the area of ​​the UAV landing area.

[0037] The second connection platform is used to transmit the attitude compensation output from the Stewart platform to the UAV landing platform. Its main functions include: expanding the UAV landing pad area; integrating a wireless charging transmitter module; providing a docking and installation reference for the UAV; maintaining the stability of the upper platform; and providing structural support for the landing platform.

[0038] In a preferred embodiment of the present invention, the positioning device can employ a Lighthouse positioning extension board. Alternatively, high-precision positioning technologies such as UWB (Ultra-Wideband) or indoor RTK can be used. The Lighthouse positioning system can consist of a SteamVR BaseStation 2.0 laser base station and a Bitcraze Lighthouse Positioning Deck (containing a photodiode array), employing an "end-side calculation" architecture: the optical drum inside the base station rotates at high speed, sequentially emitting laser pulses in both horizontal and vertical scanning planes; the positioning extension board records the time difference between the arrival of two laser pulses, calculating the horizontal and vertical angles of the object being positioned relative to the base station; after deploying two base stations with calibrated positions and attitudes in space, the system calculates the three-dimensional absolute coordinates of the object being positioned through the spatial intersection of two sets of known directional rays. In this invention, a Lighthouse Positioning Deck is installed on both the Crazyflie 2.1 UAV and the Unitree Go2 quadruped robot. Both can obtain real-time high-precision absolute positions in the same base station coordinate system, thereby directly calculating the precise relative pose of the UAV relative to the robot's landing platform. Position calculation is performed at the edge, eliminating feedback delay and enabling low-latency real-time pose output (positioning accuracy better than ±5mm, update frequency ≥100Hz).

[0039] In one embodiment of the present invention, in order for the UAV to closely follow the quadruped robot and complete its landing, it needs to be able to lock onto the quadruped robot in real time and determine the landing position. Since the Unitree Go2's position continuously changes during movement, the present invention provides a real-time dynamic target point mapping method based on a local body coordinate system. Specifically: taking the current pose of the moving subject as the origin, a local body coordinate system and a fixed offset of the landing platform in this local body coordinate system are determined; based on the real-time attitude of the moving subject, a corresponding rotation matrix is ​​determined; based on the real-time attitude, the rotation matrix, and the fixed offset, the coordinates of the landing platform in the global coordinate system are determined. Wherein, the fixed offset ( dx , dy , dz The forward offset Δx can be determined by the hardware installation location and is a known constant. 0.10 m. Real-time attitude can be published in real time via the Lighthouse positioning extension board or the moving subject's own IMU, and can be represented by a quaternion q=( q x , q y , q z , q wThis method can still accurately calculate the global spatial position of the landing platform in real time even under complex dynamic conditions such as robot dog turning and climbing.

[0040] In one embodiment, the coordinates of the landing platform in the global coordinate system can be determined according to the following formula: in,( x world , y world , z world ( ) represents the coordinates of the landing platform in the global coordinate system; x c , y c , z c R(q) represents the coordinates of the moving subject in the global coordinate system; R(q) is the rotation matrix; q is the real-time attitude; dx , dy , dz ) represents a fixed offset.

[0041] The system for drone landing and charging in dynamic scenarios provided by the present invention may further include a controller configured to execute the following program: In response to the drone's landing request, the system verifies the pose data of the mobile entity and the horizontal state of the landing platform. If the verification passes, the system acquires the positions of the drone and the mobile entity, calculates the horizontal deviation, generates a tracking control command, and drives the drone to follow the horizontally moving landing target point. If the horizontal deviation is less than a set threshold, the system drives the drone to initiate a vertical descent procedure and corrects the landing altitude target value in real time. If the distance between the drone and the landing platform reaches the determined landing altitude target value, a landing command is sent to make the drone contact the landing platform. Based on the contact information between the drone and the landing platform, the system drives the drone to automatically lock its motors and automatically activates the wireless charging module.

[0042] The Unitree Go2 exhibits real-time fluctuations in its back height during obstacle crossing and hill climbing. This invention employs a dynamic landing height compensation mechanism: it dynamically corrects the landing height based on the platform height fed back by the robot dog in real time, and reserves a preset safe buffer landing distance (e.g., 0.05 m), effectively avoiding hard landings or premature ground contact caused by the platform's dynamic movement.

[0043] In practical applications, if the status check fails or the landing times out, a forced landing process will be triggered to ensure safety.

[0044] In this invention, the above-mentioned algorithm for autonomous landing of UAVs can run in parallel with the visual recognition system (Intel RealSense D435i depth camera + YOLOv11). Under sufficient lighting and close distance (3-5 m) conditions, visual assistance can be selectively superimposed to enhance the positioning robustness. However, the algorithm provided by this invention is independent and complete and does not depend on the visual module.

[0045] This invention organically integrates multiple technological elements into a system, mainly including: 1. Mechanical-control integration: The Stewart platform and ESP32 control chip are integrated in the same structural layer, and the connection to platform A provides vibration reduction and isolation; the dual data source fusion of IMU data (feedforward compensation) from the Unitree Go2 SDK and IMU data (feedback verification) from the Stewart platform itself realizes a dual closed-loop control strategy of "robot body posture feedforward + platform posture closed loop"; 2. Three-system time-series coordination: Lighthouse positioning system guides Crazyflie 2.1 to land accurately → Stewart platform dynamic compensation ensures landing success rate → wireless charging module automatically starts to replenish power, forming a complete "flight-recovery-charging-reboot" closed-loop mission link; 3. Energy topology integration: The ESP32 (5 V) and Lighthouse expansion board are powered by the expansion interface of the Unitree Go2 body, the servo motor is powered by an independent external voltage regulator module, the wireless charging transmitter is powered by an independent power supply, and the electrical topology design effectively isolates the interference of each module.

[0046] Specifically, the technical solutions include the following: I. Overall Integration Scheme for Self-Balancing UAV Take-Off and Landing Platform. A quadruped robot-borne self-balancing UAV take-off, landing, and charging platform comprises a modular, layered integrated architecture consisting of a first connecting platform, an attitude compensation platform (a six-DOF Stewart parallel platform), a second connecting platform, a UAV landing platform, and a wireless charging module. Each layer forms a bottom-up stack on the quadruped robot's back, achieving a complete functional chain from gait disturbance compensation to landing platform stabilization to contactless power replenishment. The landing platform's limiting groove is designed according to the target rotorcraft UAV's diagonal wheelbase, achieving landing position limitation and anti-deviation functions.

[0047] 2. Autonomous recovery algorithm for UAVs based on Lighthouse laser positioning. An autonomous recovery control method running on a micro rotor UAV flight control processor and working in conjunction with a laser positioning extension board is used to guide the UAV to land on a dynamic compensation platform on the back of a mobile quadruped robot. The method may include the following steps: (1) Installing laser positioning extension boards on both the UAV and the quadruped robot, so that both can obtain three-dimensional absolute pose in real time under the same base station coordinate system, with a positioning accuracy better than ±5 mm and an update frequency of not less than 100 Hz; (2) Constructing a local body coordinate system based on the real-time pose of the quadruped robot, and mapping the fixed local offset of the landing platform center to a dynamic target point in the global coordinate system in real time through a quaternion rotation matrix, so that the target point remains effective under conditions such as robot turning and climbing; (3) Dynamically correcting the descent height target value based on the real-time feedback of the back platform height of the quadruped robot, and reserving a safety buffer distance of not less than 0.05 m to avoid hard landing; (4) Executing a complete closed-loop timing sequence of state verification → dynamic horizontal following → vertical descent → buffer contact → motor locking → charging start, with timeout protection provided for each stage. The technical effect of the method is that, in a GPS-denied, visually unmarked environment, the horizontal alignment error of the UAV relative to the mobile platform does not exceed ±10 mm.

[0048] Third, a contactless wireless charging integration solution for outdoor mobile environments. The receiver of the wireless charging module adopts the existing open-source solution of Qi WPC v1.2 based on TI BQ51013B (refer to the Bitcraze AB public schematic diagram). Specifically, the following technical solutions are adopted: (1) embedding the Qi transmitting coil into the dynamic landing platform and rigidly connecting it with the Stewart moving platform to realize the structural integration solution of attitude compensation and coil alignment synchronization; (2) a charging and discharging state machine control solution that triggers automatic charging start with landing contact signal and triggers Crazyflie 2.1 flight control unlock with CHG pin full state; (3) an energy topology design that independently powers the wireless charging transmitter and electrically isolates it from the Stewart control system.

[0049] The fourth and fifth layer modular layered integrated architecture includes a system collaboration scheme. The Stewart platform serves as the core dynamic stabilization layer of this architecture, working in tandem with the Lighthouse recovery algorithm and wireless charging module to achieve three-system timing coordination: before the landing command is triggered, the Stewart platform's horizontal compensation status is checked to ensure platform readiness before initiating the drone descent sequence; the Stewart platform's compensation actions are synchronized with the Qi transmitter's movement to ensure dynamic alignment; after landing, Crazyflie 2.1 automatically locks the drone, starts charging, and unlocks it when fully charged, forming a complete closed loop. This solves the coupling problem of dynamic platform tilt, contactless charging, and GPS rejection.

[0050] In one specific embodiment of the present invention, the mobile subject can be a Unitree Go2 EDU quadruped robot, and the specific parameters are shown in Table 1. The drone can be a Bitcraze Crazyflie 2.1 Brushless, and the specific parameters are shown in Table 2. The positioning device can be a Bitcraze Lighthouse positioning extension board, and the specific parameters are shown in Table 3. The specific parameters of the receiver of the wireless charging module are shown in Table 4.

[0051] Table 1 Table 2 Table 3 Table 4 Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention. Clearly, those skilled in the art can make various alterations and modifications to the invention without departing from its spirit and scope. Thus, if these modifications and modifications of the invention fall within the scope of the claims and their equivalents, the invention is also intended to include these modifications and modifications.

Claims

1. A system for landing and charging a drone in a dynamic scenario, characterized in that, include: The system includes a mobile main body and an attitude compensation platform, a landing platform, and a wireless charging module located on the mobile main body. The attitude compensation platform is rigidly connected to the landing platform and is used to compensate for the attitude of the mobile main body, keeping the landing platform level. The landing platform provides a landing area for the drone. A positioning device is installed on the landing platform to guide the drone to locate the mobile main body. The transmitter of the wireless charging module is embedded in the landing platform, and the receiver of the wireless charging module is installed on the bottom of the drone. When the drone lands on the landing platform, the transmitter and receiver wirelessly connect and charge the drone. When the drone finishes charging, the charging ends and the drone is unlocked.

2. The system for drone landing and charging in dynamic scenarios as described in claim 1, characterized in that, The positioning device uses the following method to guide the UAV to locate the moving subject: Using the current pose of the moving body as the origin, determine the local body coordinate system and the fixed offset of the landing platform in the local body coordinate system; Based on the real-time attitude of the moving subject, the corresponding rotation matrix is ​​determined; The coordinates of the landing platform in the global coordinate system are determined based on the real-time attitude, rotation matrix, and fixed offset.

3. The system for drone landing and charging in dynamic scenarios as described in claim 2, characterized in that, The coordinates of the landing platform in the global coordinate system are determined using the following formula: in,( x world , y world , z world ( ) represents the coordinates of the landing platform in the global coordinate system; x c , y c , z c R(q) represents the coordinates of the moving subject in the global coordinate system; R(q) is the rotation matrix; q is the real-time attitude; dx , dy , dz ) represents a fixed offset.

4. The system for landing and charging a drone in a dynamic scenario as described in claim 1, characterized in that, The system also includes a controller configured to execute the following program: In response to the drone's landing request, the pose data of the mobile subject and the horizontal state of the landing platform are verified; if the verification passes, the positions of the drone and the mobile subject are obtained, the horizontal deviation is calculated, a tracking control command is generated, and the drone is driven to follow the horizontally moving landing target point. If the horizontal deviation is less than the set threshold, the drone will initiate a vertical descent procedure and the target descent altitude will be corrected in real time. If the distance between the drone and the landing platform reaches the determined target descent altitude, a descent command will be sent to make the drone contact the landing platform. Based on the contact information between the drone and the landing platform, the drone will automatically lock its motors and the wireless charging module will automatically start.

5. The system for landing and charging a drone in a dynamic scenario as described in claim 1, characterized in that, The system also includes a first connection platform located between the mobile body and the attitude compensation platform, which is used to achieve geometric alignment, stiffness transition and vibration isolation between the installation interface of the mobile body and the base of the attitude compensation platform.

6. The system for landing and charging a drone in a dynamic scenario as described in claim 1, characterized in that, The system also includes a second connection platform, which is located between the attitude compensation platform and the landing platform, and is used to receive the output of the attitude compensation platform and expand the area of ​​the UAV landing area.

7. The system for landing and charging a drone in a dynamic scenario as described in claim 1, characterized in that, The compensation platform is a six-degree-of-freedom parallel self-stabilizing platform used to compensate for the motion of the moving body and the pitch and roll disturbances caused by the undulations of the ground in real time, so that the landing platform always maintains a dynamic horizontal state.

8. The system for landing and charging a drone in a dynamic scenario as described in claim 1, characterized in that, The transmitter of the wireless charging module uses an independent power supply, which is isolated from the electrical system of the mobile device.

9. The system for landing and charging a drone in a dynamic scenario as described in claim 1, characterized in that, The mobile body is a quadruped robot.

10. The system for landing and charging a drone in a dynamic scenario as described in claim 1, characterized in that, The positioning device uses one of the following: Lighthouse positioning expansion board, UWB ultra-wideband, or indoor RTK.

Images