Vehicle-mounted unmanned aerial vehicle flight control method and vehicle-mounted unmanned aerial vehicle flight control system

Abstract

The invention provides a vehicle-mounted unmanned aerial vehicle flight control method and a vehicle-mounted unmanned aerial vehicle flight control system. The method comprises the steps that a first return flight track of the unmanned aerial vehicle flying to take-off and landing equipment is determined according to the position of the unmanned aerial vehicle and the position of the take-off and landing equipment matched with the unmanned aerial vehicle, the unmanned aerial vehicle is controlled to fly according to the first return flight track, the take-off and landing equipment is arranged on a vehicle, and the vehicle is in a moving state; when the unmanned aerial vehicle flies according to the first return flight trajectory, if a spherical marker on the take-off and landing equipment is detected, determining a second return flight trajectory of the unmanned aerial vehicle flying to the take-off and landing equipment according to the position of the unmanned aerial vehicle and the position of the spherical marker, and controlling the unmanned aerial vehicle to fly according to the second return flight trajectory; when the unmanned aerial vehicle flies according to the second return flight track, if a visual positioning mark on the take-off and landing equipment is detected, the unmanned aerial vehicle is controlled to land on the take-off and landing equipment according to the position of the unmanned aerial vehicle and the position of the visual positioning mark. According to the invention, accurate landing of the unmanned aerial vehicle can be realized on the mobile take-off and landing equipment.

Description

Technical Field

[0001] This application relates to the field of unmanned aerial vehicle (UAV) control technology, and in particular to a vehicle-mounted UAV flight control method and a vehicle-mounted UAV flight control system. Background Technology

[0002] In recent years, drone technology has made significant progress and has been widely used in various fields such as inspection, logistics delivery, and emergency response. Therefore, achieving autonomous take-off and landing of drones on mobile platforms such as vehicles has become an important direction in the current research field of drone technology.

[0003] However, existing vehicle-mounted drone take-off and landing technologies have at least the following problems: Traditional drones mainly rely on a single Global Positioning System (GPS) or simple planar guide signs for positioning. When the drone has a large observation angle relative to the guide sign, the accuracy and reliability of the guide sign recognition cannot be fully guaranteed. This makes it difficult for the drone to accurately perceive its relative position to the fast and complex moving vehicle platform, thus it cannot effectively adapt to various changes in the vehicle platform during movement and cannot land accurately on the vehicle platform. Summary of the Invention

[0004] This application provides a vehicle-mounted UAV flight control method and a vehicle-mounted UAV flight control system to solve the shortcomings of existing technologies where UAVs cannot effectively adapt to various changes in the vehicle platform during movement and cannot land accurately on the vehicle platform. This application uses a combination of spherical markers and visual positioning marks for guidance, enabling the UAV to adapt to various changes in the vehicle platform (i.e., take-off and landing equipment) during movement and ultimately achieve accurate landing on the take-off and landing equipment.

[0005] This application provides a vehicle-mounted drone flight control method, applied to a flight controller in a drone, the vehicle-mounted drone flight control method comprising: When it is determined that a return-to-home operation will be performed on the drone, a first return-to-home trajectory for the drone to fly toward the take-off and landing equipment is determined based on the location of the drone and the location of the take-off and landing equipment adapted to the drone, and the drone is controlled to fly according to the first return-to-home trajectory. The take-off and landing equipment is mounted on a vehicle and the vehicle is in motion. During the flight of the UAV along the first return trajectory, if a spherical marker is detected on the take-off and landing equipment, a second return trajectory for the UAV to fly toward the take-off and landing equipment is determined based on the position of the UAV and the position of the detected spherical marker, and the flight of the UAV is controlled according to the second return trajectory; During the flight of the UAV along the second return trajectory, if a visual positioning marker is detected on the take-off and landing equipment, the UAV is controlled to land on the take-off and landing equipment based on the position of the UAV and the position of the detected visual positioning marker.

[0006] According to the vehicle-mounted unmanned aerial vehicle (UAV) flight control method of this application, determining the first return trajectory of the UAV towards the take-off and landing equipment based on the position of the UAV and the position of the take-off and landing equipment adapted to the UAV includes: The dynamic status data of the UAV and the dynamic status data of the take-off and landing equipment are obtained. The dynamic status data of the UAV describes the motion law of the UAV, and the dynamic status data of the take-off and landing equipment describes the motion law of the take-off and landing equipment. Based on the location of the UAV, the location of the take-off and landing equipment, the dynamic status data of the UAV, and the dynamic status data of the take-off and landing equipment, the first return trajectory is determined by an extended Kalman filter algorithm.

[0007] According to the vehicle-mounted unmanned aerial vehicle (UAV) flight control method of this application, determining the second return trajectory of the UAV towards the take-off and landing equipment based on the position of the UAV and the detected position of the spherical marker includes: Acquire the dynamic status data of the UAV and the dynamic status data of the take-off and landing equipment; Based on the location of the UAV, the location of the spherical marker, the dynamic status data of the UAV, and the dynamic status data of the take-off and landing equipment, the second return trajectory is determined using an extended Kalman filter algorithm.

[0008] According to the vehicle-mounted unmanned aerial vehicle (UAV) flight control method of this application, the step of controlling the UAV to land on the take-off and landing equipment based on the UAV's position and the detected position of the visual positioning marker includes: With the goal of aligning the center point of the UAV with the center point of the visual positioning mark, and controlling the nose direction of the UAV to be consistent with the movement direction of the take-off and landing equipment, the actions of the UAV are adjusted according to the position of the UAV and the position of the visual positioning mark. After the adjustment is completed, the drone is controlled to land on the take-off and landing equipment.

[0009] According to the vehicle-mounted unmanned aerial vehicle (UAV) flight control method of this application, the step of adjusting the UAV's movements based on the position of the UAV and the position of the visual positioning mark, with the goal of aligning the center point of the UAV with the center point of the visual positioning mark and controlling the UAV's nose direction to be consistent with the movement direction of the take-off and landing equipment, includes: Based on the position of the UAV and the position of the visual positioning marker, the lateral drift error of the UAV is determined, and the actions of the UAV are adjusted according to the lateral drift error until the lateral drift error is less than a first error threshold. The yaw angle error of the UAV is determined based on the deviation between the UAV's nose direction and the direction of movement of the take-off and landing equipment, and the UAV's actions are adjusted according to the yaw angle error until the yaw angle error is less than a second error threshold.

[0010] According to the vehicle-mounted unmanned aerial vehicle (UAV) flight control method of this application, controlling the UAV to land on the take-off and landing equipment includes: If the distance between the drone and the take-off and landing equipment is less than a preset height threshold, the descent speed of the drone is determined based on the distance between the drone and the take-off and landing equipment. Based on the descent speed, the drone is controlled to land on the take-off and landing equipment.

[0011] According to the vehicle-mounted unmanned aerial vehicle (UAV) flight control method of this application, controlling the UAV to land on the take-off and landing equipment includes: During the process of the drone landing on the take-off and landing equipment, first disturbance information caused by the take-off and landing equipment to the drone and second disturbance information caused by the environment in which the drone is located to the drone are acquired. The actions of the UAV are adjusted with the goal of eliminating the first and second disturbance information.

[0012] This application also provides a vehicle-mounted unmanned aerial vehicle (UAV) flight control system, including ground control equipment, UAV, and take-off and landing equipment, wherein the ground control equipment is communicatively connected to the UAV and the take-off and landing equipment respectively. The ground control equipment is used to control the operating status of the UAV and the take-off and landing equipment; The take-off and landing equipment has a built-in square take-off and landing module and a fastening module. The surface of the square take-off and landing module is provided with visual identification marks. The fastening modules are respectively located at the four corners of the square take-off and landing module. The top of the fastening module is provided with a spherical marker. The take-off and landing equipment is used to provide take-off and landing sites to the UAV under the control of the ground control equipment, and to send its own position to the UAV. The take-off and landing equipment is mounted on a vehicle. The drone is used to perform flight missions under the control of the ground control equipment. The drone has a built-in flight controller, which is used to determine the return trajectory based on the drone's position, the position of the spherical marker, the position of the visual recognition mark, and the position of the take-off and landing equipment during the drone's flight, and to control the drone to land on the take-off and landing equipment based on the return trajectory.

[0013] This application also provides a flight controller for a drone, including a memory, a processor, and a computer program stored in the memory and executable on the processor. When the processor executes the computer program, it implements the vehicle-mounted drone flight control method described above.

[0014] This application also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the vehicle-mounted unmanned aerial vehicle flight control method as described above.

[0015] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the vehicle-mounted unmanned aerial vehicle flight control method as described above.

[0016] The method of this application, when the UAV is far from the take-off and landing equipment and cannot detect the spherical marker on the take-off and landing equipment, the flight controller automatically plans a first return trajectory for the UAV to fly to the take-off and landing equipment. When the UAV is close to the take-off and landing equipment and can detect the spherical marker on the take-off and landing equipment, but cannot yet detect the visual positioning mark on the take-off and landing equipment (when the UAV has a large observation angle relative to the take-off and landing equipment, the spherical marker can be detected first), the flight controller plans a second return trajectory for the UAV to fly to the take-off and landing equipment based on the position of the UAV and the position of the detected spherical marker. When the UAV is close enough to the take-off and landing equipment and can detect the visual positioning mark on the take-off and landing equipment, the flight controller controls the UAV to land on the take-off and landing equipment based on the position of the UAV and the position of the detected visual positioning mark. This application employs a combination of spherical markers and visual positioning marks for guidance. Since the spherical markers can be accurately detected even when the UAV has a large viewing angle relative to the take-off and landing equipment, the accuracy and reliability of the guidance marks are guaranteed. This enables the UAV to accurately perceive its relative position to the fast-moving and complex vehicle platform (i.e., the take-off and landing equipment), and effectively adapt to various changes in the vehicle platform during its movement, thereby allowing it to land precisely on the take-off and landing equipment. Attached Figure Description

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

[0018] Figure 1 This is a flowchart illustrating a flight control method for a vehicle-mounted unmanned aerial vehicle (UAV) according to an embodiment of this application; Figure 2 This is a schematic diagram of a vehicle-mounted unmanned aerial vehicle (UAV) flight control system according to an embodiment of this application; Figure 3 This is a schematic diagram of the structure of a flight controller shown in one embodiment of this application. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0020] The vehicle-mounted UAV flight control method of this application is applied to the flight controller in the UAV. Figure 1 This is a flowchart illustrating a flight control method for a vehicle-mounted unmanned aerial vehicle (UAV) according to an embodiment of this application. (Refer to...) Figure 1 The vehicle-mounted unmanned aerial vehicle flight control method of this application may include: Step 101: When it is determined that a return-to-home operation will be performed on the drone, the first return-to-home trajectory of the drone to the take-off and landing equipment is determined based on the location of the drone and the location of the take-off and landing equipment adapted to the drone. The drone is controlled to fly according to the first return-to-home trajectory. The take-off and landing equipment is set on the vehicle and the vehicle is in motion.

[0021] In this application, when the UAV needs to return after completing its flight mission, the flight controller obtains the current position of the UAV and the current position of the take-off and landing equipment, and automatically plans the first return trajectory for the UAV to fly towards the take-off and landing equipment. Then, the flight controller controls the UAV to fly along the first return trajectory to approach the take-off and landing equipment.

[0022] In this application, the take-off and landing equipment actively sends its real-time position to the flight controller so that the flight controller can perform trajectory planning.

[0023] The take-off and landing equipment is equipped with a first real-time kinematic (RTK) base station to obtain the real-time position of the take-off and landing equipment. The UAV is equipped with a second real-time kinematic base station to obtain the real-time position of the UAV.

[0024] The RTK base stations used in this application are all RTK-GNSS (Real Time Kinematic-Global Navigation Satellite System Base Stations), which can receive satellite signals from multiple global navigation satellite systems, thereby obtaining the precise location of UAVs and take-off and landing equipment.

[0025] Step 102: During the flight of the UAV along the first return trajectory, if a spherical marker is detected on the take-off and landing equipment, determine the second return trajectory of the UAV to the take-off and landing equipment based on the position of the UAV and the position of the detected spherical marker, and control the flight of the UAV according to the second return trajectory.

[0026] In this application, when the UAV is flying along the first return trajectory, the flight controller estimates the distance between the UAV and the take-off and landing equipment in real time. If the distance is less than a preset first distance threshold, it enters the spherical marker recognition mode. In the spherical marker recognition mode, the UAV's downward-facing camera acquires ground images in real time, and the acquired images are analyzed by the UAV's internal vision processing module. If the image contains a spherical marker on the take-off and landing equipment, it is determined that a spherical marker on the take-off and landing equipment has been detected.

[0027] The vision processing module can use the YOLO (You Only Look Once) V10 algorithm to improve the recognition accuracy of spherical markers when recognizing them in an image.

[0028] In this application, the vision processing module pre-stores the actual size of the spherical marker. Therefore, it can estimate the distance from the drone to the spherical marker based on the size of the spherical marker displayed in the image (e.g., how many pixels its diameter occupies) and parameters such as the focal length of the downward-facing camera, using perspective principles. By combining the center position of the spherical marker in the image and the orientation of the downward-facing camera, the three-dimensional position of the spherical marker can be estimated.

[0029] Specifically, the vision processing module determines the direction vector of the spherical marker in the downward-facing camera coordinate system based on the center position of the spherical marker in the image and the internal parameters of the downward-facing camera (such as focal length). Next, according to the orientation of the downward-facing camera, this direction vector is transformed into the UAV's body coordinate system or the global coordinate system. Simultaneously, based on the size of the spherical marker in the image (e.g., how many pixels its diameter occupies) and its known actual dimensions, combined with the internal parameters of the downward-facing camera, the distance from the UAV (downward-facing camera) to the spherical marker is estimated using perspective principles. Finally, combining the direction vector originating from the downward-facing camera and the estimated distance, the precise position of the spherical marker in three-dimensional space (i.e., its three-dimensional coordinates relative to the downward-facing camera or the UAV) is calculated.

[0030] Therefore, the flight controller can continue to plan the second return trajectory of the drone towards the take-off and landing equipment based on the current position of the drone and the three-dimensional position of the detected spherical marker, and control the drone's flight according to the second return trajectory to further approach the take-off and landing equipment.

[0031] Step 103: During the flight of the UAV along the second return trajectory, if a visual positioning marker is detected on the take-off and landing equipment, control the UAV to land on the take-off and landing equipment based on the position of the UAV and the position of the detected visual positioning marker.

[0032] In this application, when the UAV flies along the second return trajectory, the flight controller estimates the distance between the UAV and the take-off and landing equipment in real time. If this distance is less than a preset second distance threshold, it enters the visual positioning mark recognition mode. In the visual positioning mark recognition mode, the visual processing module analyzes the images captured in real time by the downward-facing camera. If the image contains visual positioning marks on the take-off and landing equipment, it is determined that visual positioning marks on the take-off and landing equipment have been detected. The second distance threshold is less than the first distance threshold. The first and second distance thresholds can be set according to actual needs.

[0033] In this application, the visual positioning marker can be a two-dimensional planar marker. Specifically, the visual positioning marker can be a QR code or an ArUco code (Augmented Reality University of Coimbra Marker), which can be set according to actual needs.

[0034] In this application, the vision processing module pre-stores the actual physical dimensions (e.g., side length) of the visual positioning markers. The vision processing module processes the images captured by the downward-facing camera and detects and identifies the visual positioning markers in the images using a specific algorithm.

[0035] Specifically, the algorithm precisely extracts the key feature points of the visual positioning markers in the image. Since the structure of the visual positioning markers is known, these feature points in the image can be linked to their corresponding 3D coordinates in the actual physical world (for example, knowing which pixel in the image corresponds to the top-left corner of the visual positioning marker, and where that corner is located in the physical world). Next, based on the known 3D feature point coordinates of the visual positioning markers in the physical world (e.g., the actual physical positions of the four corners of the visual positioning markers, where the relative relationships are fixed), the pixel coordinates of the feature points captured by the downward-facing camera in the 2D image, and the internal parameters of the downward-facing camera (such as focal length and distortion), the algorithm calculates a 3D transformation describing the position and orientation of the visual positioning markers relative to the downward-facing camera in 3D space (including 3D translation and 3D rotation).

[0036] Next, the flight controller can control the drone to land on the take-off and landing equipment based on the position and attitude of the visual positioning marker relative to the downward-looking camera in three-dimensional space, combined with the drone's own real-time position.

[0037] The method of this application, when the UAV is far from the take-off and landing equipment and cannot detect the spherical marker on the take-off and landing equipment, the flight controller automatically plans a first return trajectory for the UAV to fly to the take-off and landing equipment. When the UAV is close to the take-off and landing equipment and can detect the spherical marker on the take-off and landing equipment, but cannot yet detect the visual positioning mark on the take-off and landing equipment (when the UAV has a large observation angle relative to the take-off and landing equipment, the spherical marker can be detected first), the flight controller plans a second return trajectory for the UAV to fly to the take-off and landing equipment based on the position of the UAV and the position of the detected spherical marker. When the UAV is close enough to the take-off and landing equipment and can detect the visual positioning mark on the take-off and landing equipment, the flight controller controls the UAV to land on the take-off and landing equipment based on the position of the UAV and the position of the detected visual positioning mark. This application employs a combination of spherical markers and visual positioning marks for guidance. Since the spherical markers can be accurately detected even when the UAV has a large viewing angle relative to the take-off and landing equipment, the accuracy and reliability of the guidance marks are guaranteed. This enables the UAV to accurately perceive its relative position to the fast-moving and complex vehicle platform (i.e., the take-off and landing equipment), effectively adapting to various changes in the vehicle platform during its movement. As a result, it can accurately land on the take-off and landing equipment, ensuring the stability and safety of the return landing.

[0038] In conjunction with the above embodiments, in one implementation, step 101 may include: Acquire dynamic status data of the UAV and the take-off and landing equipment. The dynamic status data of the UAV describes the motion law of the UAV, and the dynamic status data of the take-off and landing equipment describes the motion law of the take-off and landing equipment. Based on the location of the UAV, the location of the take-off and landing equipment, the dynamic status data of the UAV, and the dynamic status data of the take-off and landing equipment, the first return trajectory is determined by using the extended Kalman filter algorithm.

[0039] In this application, the take-off and landing equipment is equipped with a first inertial measurement unit (IMU). The first inertial measurement unit can collect dynamic state data of the take-off and landing equipment in real time. The dynamic state data of the take-off and landing equipment mainly includes velocity, acceleration, heading angle, and roll angle (including roll and pitch angles). The heading angle refers to the orientation of the take-off and landing equipment on the horizontal plane (the take-off and landing equipment follows the vehicle's movement). The roll angle refers to the degree of tilt of the take-off and landing equipment relative to the horizontal plane. The roll angle refers to the rotation angle of the take-off and landing equipment about its forward direction axis, reflecting the degree of left and right tilt of the vehicle. The pitch angle refers to the rotation angle of the take-off and landing equipment about its left and right direction axes, reflecting the degree of nose-up or nose-down of the vehicle (e.g., uphill or downhill).

[0040] The drone is equipped with a second inertial measurement module. This second inertial measurement module can collect the drone's dynamic state data in real time. The drone's dynamic state data mainly includes the drone's acceleration along three axes (usually forward / backward, left / right, up / down) in the body coordinate system, and the drone's rotational angular velocity around the three axes (corresponding to roll, pitch, and yaw) in the body coordinate system.

[0041] In this application, the Extended Kalman Filter (EKF) is a recursive filter used to estimate the state of a dynamic nonlinear system in the presence of noise. It approximates the nonlinear system as a linear system through linearization, and then applies the prediction and update steps of the standard Kalman filter. The EKF algorithm predicts the future state of the system based on its mathematical model and noisy measurement data, and corrects the prediction based on new measurement data to obtain the optimal estimate of the system state.

[0042] The take-off and landing equipment obtains its own dynamic status data in real time through the second inertial measurement module and transmits this data to the UAV's flight controller. The flight controller can accurately determine the first return trajectory based on the UAV's current position, the take-off and landing equipment's current position, the UAV's real-time dynamic status data, and the take-off and landing equipment's real-time dynamic status data, using an extended Kalman filter (EKF) algorithm. In this application, since both the take-off and landing equipment and the UAV are in a state of dynamic change, the position data collected by the RTK-GNSS base station and the dynamic status data collected by the inertial measurement module are fused using the EKF algorithm. This allows the dynamic status data to compensate for errors caused by the dynamic changes of the UAV and the take-off and landing equipment, thereby planning a more suitable return trajectory for the UAV and ensuring the accuracy of the UAV landing.

[0043] In conjunction with the above embodiments, in one implementation, step 102 may include: Acquire dynamic status data of the drone and its take-off and landing equipment; Based on the location of the UAV, the location of the spherical marker, the dynamic status data of the UAV, and the dynamic status data of the take-off and landing equipment, the second return trajectory is determined by using the extended Kalman filter algorithm.

[0044] In this application, when the UAV is close to the take-off and landing equipment and can detect the spherical marker on the take-off and landing equipment but cannot detect the visual positioning mark on the take-off and landing equipment, the flight controller plans a suitable second return trajectory for the UAV based on the current position of the UAV, the current position of the spherical marker, the current dynamic status data of the UAV, and the current dynamic status data of the take-off and landing equipment, using an extended Kalman filter algorithm, thereby ensuring the accuracy of the UAV landing.

[0045] In conjunction with the above embodiments, in one implementation, step 103 may include: With the goal of aligning the center point of the drone with the center point of the visual positioning mark, and controlling the direction of the drone's nose to be consistent with the direction of movement of the take-off and landing equipment, the drone's movements are adjusted according to the position of the drone and the position of the visual positioning mark. After completing the adjustments, control the drone to land on the take-off and landing equipment.

[0046] In this application, when the UAV is close enough to the take-off and landing equipment (approximately 10 meters above the take-off and landing equipment) and the UAV can detect the visual positioning mark on the take-off and landing equipment, the flight controller controls the UAV's movements based on the position of the UAV and the position of the visual positioning mark, so that the center point of the UAV is aligned with the center point of the visual positioning mark, and the direction of the UAV's nose is consistent with the direction of movement of the take-off and landing equipment.

[0047] The alignment of the UAV's center point with the visual positioning marker's center point means that the deviation between the UAV's center point and the visual positioning marker's center point is less than a preset deviation. The alignment of the UAV's nose direction with the direction of motion of the take-off and landing equipment means that the angle between the UAV's nose direction and the direction of motion of the take-off and landing equipment is less than a preset angle.

[0048] Once the center point of the drone is aligned with the center point of the visual positioning mark and the direction of the drone's nose is consistent with the direction of movement of the take-off and landing equipment, the flight controller controls the drone to descend until it lands on the take-off and landing equipment.

[0049] In this application, the center point of the UAV is first aligned with the center point of the visual positioning mark, and the direction of the UAV's nose is aligned with the direction of movement of the take-off and landing equipment. Then, the UAV is controlled to land, which can ensure that the UAV can land accurately on the take-off and landing equipment that is in motion.

[0050] In one implementation, based on the above embodiments, with the goal of aligning the center point of the drone with the center point of the visual positioning mark, and controlling the drone's nose direction to be consistent with the movement direction of the take-off and landing equipment, the drone's movements are adjusted according to the drone's position and the position of the visual positioning mark. Specifically, this may include: Based on the location of the drone and the location of the visual positioning marker, the lateral drift error of the drone is determined, and the drone's actions are adjusted according to the lateral drift error until the lateral drift error is less than the first error threshold. The yaw angle error of the drone is determined based on the deviation between the drone's nose direction and the direction of movement of the take-off and landing equipment. The drone's actions are then adjusted based on the yaw angle error until the yaw angle error is less than the second error threshold.

[0051] In this application, lateral drift error is used express. This indicates the positional deviation or offset of the UAV's center point relative to the center point of the visual positioning marker along the X-axis. =X 无人机 -X 视觉定位标记 . This represents the positional deviation or offset of the drone relative to the center point of the visual positioning marker along the Y-axis. = Y 无人机 -Y 视觉定位标记 .

[0052] (X 视觉定位标记 Y 视觉定位标记 (X) represents the position coordinates of the center point of the visual positioning marker in a two-dimensional plane coordinate system. 无人机 Y 无人机The coordinates represent the position coordinates of the UAV in a two-dimensional plane coordinate system. This two-dimensional plane coordinate system is pre-established on the take-off and landing equipment, and the center point of the visual positioning mark can be used as the origin or reference point.

[0053] In this application, the first error threshold and the second error threshold can be set according to actual needs.

[0054] In this application, after determining After that, it can be based on Generate specific motion adjustment instructions, and adjust the drone's motion according to these instructions until... The size value is less than the first error threshold (e.g., 5cm) and The value is less than the first error threshold (e.g., 5cm).

[0055] In this application, yaw angle refers to the rotation of the UAV around its own vertical axis, that is, the orientation or direction of the UAV. Yaw angle error refers to the difference between the actual yaw angle of the UAV and the expected yaw angle (or orientation). In vehicle-mounted take-off and landing scenarios, for the convenience of subsequent automatic fixing and charging, it is usually necessary to ensure that the nose direction of the UAV is consistent with the forward direction of the vehicle (take-off and landing equipment) when landing.

[0056] In this application, the deviation between the drone's nose orientation and the direction of motion of the take-off and landing equipment can be determined based on the information of the identified visual positioning marker. The pose of the identified visual positioning marker relative to the downward-facing camera includes the rotation information of the visual positioning marker relative to the coordinate system of the downward-facing camera. By analyzing this rotation information, the rotation angle of the visual positioning marker relative to the downward-facing camera in three-dimensional space on each axis can be determined, including the rotation around the vertical axis, i.e., the yaw angle. Since the attitude of the downward-facing camera relative to the drone body is known and fixed, and the visual positioning marker of the take-off and landing equipment is set to be aligned with the vehicle's forward direction (the direction of motion of the take-off and landing equipment is consistent with the direction of motion of the vehicle), the yaw angle of the visual positioning marker relative to the downward-facing camera calculated by the visual processing module, after coordinate system transformation, can reflect the difference in yaw angle between the drone body and the forward direction of the take-off and landing equipment. The difference in yaw angle is the deviation between the drone's nose orientation and the direction of motion of the take-off and landing equipment.

[0057] Therefore, after determining the lateral drift error and yaw angle error, the flight controller immediately determines the adjustment actions for the drone (such as moving left or right, forward or backward, and rotating the fuselage) and further eliminates these deviations, enabling the drone to accurately align with the center point of the visual positioning mark and ensure that the nose is facing the same direction as the takeoff and landing equipment. Then, after correcting these deviations, the flight controller initiates the drone's descent.

[0058] In this application, during the descent of the UAV, the UAV continuously receives the position of the visual positioning marker processed by the vision processing module, and uses a PID (Proportional-Integral-Derivative) control algorithm to calculate the lateral drift error based on the position of the visual positioning marker and the position of the UAV. and according to Error correction is performed to ensure the drone accurately aligns with the target landing point. The mathematical expression of the PID control algorithm is as follows: ,in It is a real-time error. It is the calculated control output.

[0059] In conjunction with the above embodiments, in one implementation, controlling the drone to land on the take-off and landing equipment includes: If the drone's height from the take-off and landing equipment is less than a preset height threshold, the drone's descent speed is determined based on the drone's height from the take-off and landing equipment. Based on the descent speed, control the drone to land on the take-off and landing equipment.

[0060] In this application, the preset height threshold is a small value, which can be set according to actual needs, for example, it can be set to 2m.

[0061] When the drone is less than 2 meters above the take-off and landing equipment, the laser rangefinder inside the drone is activated and measures the vertical distance between the drone and the take-off and landing equipment below in real time.

[0062] Based on real-time altitude data measured by a laser rangefinder, the flight controller precisely adjusts the drone's vertical descent speed to achieve a smooth landing, with a descent speed range of 0.2-0.5 m / s. During descent, the flight controller employs a dynamic altitude control strategy: when the altitude between the drone and the takeoff and landing equipment is greater than 1 m, its descent speed will gradually decrease; when the altitude between the drone and the takeoff and landing equipment is less than 0.5 m, the flight controller will activate an ultra-slow descent mode, further reducing the descent speed to approximately 0.1-0.2 m / s, thereby ensuring a soft landing for the drone.

[0063] Secondly, when the drone is less than 2 meters above the takeoff and landing equipment, the flight controller will also use the PID control algorithm to calculate the lateral drift error. and according to Error correction is performed. During this process, the drone's movements are adjusted to ensure the lateral drift error is less than a third error threshold, which is less than the first error threshold. For example, the third error threshold could be 0.2 cm, thus ensuring the drone accurately aligns with the target landing point.

[0064] In this application, during the fine-tuning stage below 2m, the role of PID control becomes more critical due to the higher precision requirements. Therefore, when the UAV is at least 2m above the takeoff and landing equipment, lateral drift error correction is necessary. The frequency can be less than when the drone's distance from the takeoff and landing equipment is less than 2 meters, to correct lateral drift error. This frequency ensures efficient drone landing and saves drone power.

[0065] In conjunction with the above embodiments, in one implementation, controlling the drone to land on the take-off and landing equipment may include: During the process of the drone landing on the take-off and landing equipment, information on the first disturbance caused by the take-off and landing equipment to the drone and information on the second disturbance caused by the environment in which the drone is located to the drone are acquired. The drone's actions are adjusted with the goal of eliminating the first and second disturbance information.

[0066] In this application, during a specific phase of the UAV's descent, the flight controller performs dynamic compensation control to ensure a stable and precise landing in a dynamic and wind-affected environment. This phase mainly includes the following two aspects: (1) Dynamic attitude compensation of the platform (take-off and landing equipment) (corresponding to the first disturbance information): The flight controller monitors the dynamic status data of the takeoff and landing equipment in real time, obtaining information such as turbulence, roll angle, pitch angle, and vehicle motion (e.g., turning, acceleration, deceleration). Then, based on this data, the flight controller uses the EKF prediction algorithm to dynamically predict the future attitude and position changes of the takeoff and landing equipment. Based on the prediction results, the flight controller dynamically adjusts the UAV's attitude (including roll angle, pitch angle, and yaw angle) in real time to ensure that the UAV remains relatively parallel to the takeoff and landing equipment at the moment of contact. This compensation mechanism aims to counteract the impact of the takeoff and landing equipment's movement on the UAV's landing attitude.

[0067] (2) Wind field disturbance compensation control (corresponding to the second disturbance information): The drone monitors wind speed and direction in real time using its onboard sensors (such as airspeed indicators or IMU feedback). Based on this real-time wind data, the flight controller uses adaptive control algorithms to automatically adjust the drone's attitude (including roll, pitch, and yaw angles) to compensate for the interference caused by wind disturbances in real time. This reduces the displacement or tilting of the drone during landing due to wind effects, thus enhancing landing stability.

[0068] In this application, the attitude of a UAV refers to its orientation or posture in three-dimensional space, described using roll angle, pitch angle, and yaw angle. Roll angle is the rotation angle about the UAV's front-to-rear axis (from nose to tail), controlling the UAV's left-to-right tilt. Pitch angle is the rotation angle about the UAV's left-to-right axis (from wingtip to wingtip), controlling the UAV's pitching up or down. Yaw angle is the rotation angle about the UAV's vertical axis, controlling the UAV's heading or rotation direction.

[0069] In this application, when the drone's altitude drops to within about 10 cm of the surface of the take-off and landing equipment, the flight controller will automatically control the drone to slowly reduce the power generated by the rotor to a minimum level, so as to ensure that the drone eventually makes gentle contact with the surface of the take-off and landing equipment at an extremely low contact speed (e.g., less than 0.1 m / s), thereby achieving a soft landing.

[0070] In this application, the surface of the take-off and landing equipment is also equipped with a dedicated buffer and shock-absorbing pad, such as a rubber pad, silicone pad, or air cushion structure, so as to further absorb the residual impact energy at the moment the UAV contacts the take-off and landing equipment, significantly reduce the impact force during landing, and effectively prevent damage to the structure of the UAV and the take-off and landing equipment.

[0071] The landing process of a drone after completing its mission will be described below using a complete embodiment, which includes the following steps: Step 1: Return and Initial Positioning Phase (Remote to Approaching Takeoff and Landing Equipment). The flight controller receives dynamic status information and the location of the takeoff and landing equipment. Based on its own location, its own dynamic status information, the location of the takeoff and landing equipment, and the dynamic status information of the takeoff and landing equipment, the flight controller uses built-in logic and EKF prediction algorithm to plan a flight path (first return trajectory) to the area above the takeoff and landing equipment. The flight controller then controls the UAV to begin its return flight according to the first return trajectory.

[0072] Step 2: Approaching the Take-off and Landing Equipment and Initial Visual Recognition Phase. The flight controller guides the UAV to fly to the area above the take-off and landing equipment (usually above 10m). When the distance between the UAV and the take-off and landing equipment is less than a pre-set first distance threshold, it enters the spherical marker recognition mode. In this mode, the flight controller activates the downward-facing camera. The downward-facing camera sends the acquired image data to the vision processing module. The vision processing module uses the YOLO algorithm to process the image data and identify the spherical marker on the take-off and landing equipment. Based on the UAV's real-time position and the position of the identified spherical marker, the flight controller determines the UAV's second return trajectory towards the take-off and landing equipment and controls the UAV's flight according to the second return trajectory.

[0073] Step 3: Precise Visual Positioning and Track Correction Stage. When the distance between the UAV and the takeoff and landing equipment is less than a pre-set second distance threshold, the UAV enters the visual positioning marker recognition mode (at which point the UAV's altitude is approximately 10 meters). In this mode, the downward-facing camera captures images of the visual positioning markers. The visual processing module processes the image data and calculates the real-time relative position error of the UAV relative to the visual positioning markers. And yaw angle error. Based on the relative position error (i.e., lateral drift error) and yaw angle error, the flight controller begins to correct the trajectory, adjusting the UAV's horizontal position and orientation to align with the landing point. During this stage, dynamic compensation continues.

[0074] Step 4: Fine Control Phase (Height Below 2 Meters). When the drone's height relative to the landing gear decreases to within 2 meters, the flight controller enters fine control mode. The flight controller activates the laser rangefinder and receives real-time data on the vertical distance between the drone and the platform measured by the laser rangefinder. Based on this vertical distance data, the flight controller controls the drone's descent speed. When the height is greater than 1 meter, the descent speed is maintained within the range of 0.2~0.5 m / s and gradually decreased. When the height is less than 0.5 meters, an ultra-slow descent (approximately 0.1~0.2 m / s) is initiated. The flight controller continuously receives real-time relative position errors calculated by the vision processing module and quickly compensates for the drone's lateral drift error based on the relative position error and PID control algorithm, keeping the drone always directly above the target point (the center point of the visual positioning marker). In addition, the flight controller continuously receives real-time dynamic status information and wind field information from the landing gear, performs dynamic attitude compensation for the platform and wind disturbance compensation, and adjusts the drone's roll, pitch, and yaw attitudes to maintain stability and alignment with the target point.

[0075] Step 5: Soft Landing Implementation Phase (approximately 10 cm altitude). When the altitude drops to approximately 10 cm, the flight controller automatically and slowly reduces rotor power to a minimum based on data from the laser rangefinder. Finally, the drone gently touches the surface of the vehicle-mounted platform, achieving a soft landing.

[0076] The following section will introduce a vehicle-mounted unmanned aerial vehicle (UAV) flight control system provided in this application. Figure 2 This is a schematic diagram illustrating a vehicle-mounted unmanned aerial vehicle (UAV) flight control system according to an embodiment of this application. (Refer to...) Figure 2 The vehicle-mounted unmanned aerial vehicle (UAV) flight control system of this application includes ground control equipment, UAV, and take-off and landing equipment, with the ground control equipment communicating with both the UAV and the take-off and landing equipment.

[0077] Ground control equipment is used to control the operating status of drones and take-off and landing equipment; The take-off and landing equipment has a built-in square take-off and landing module and a fastening module. The surface of the square take-off and landing module is equipped with visual identification marks. The fastening modules are respectively set at the four corners of the square take-off and landing module. The top of the fastening module is equipped with a spherical marker. The take-off and landing equipment is used to provide take-off and landing sites for the UAV under the control of the ground control equipment, and to send its own position to the UAV. The take-off and landing equipment is mounted on a vehicle. Unmanned aerial vehicles (UAVs) are used to perform flight missions under the control of ground control equipment. The UAV has a built-in flight controller, which is used to determine the return trajectory based on the UAV's position, the position of the spherical marker, the position of the visual recognition mark, and the position of the take-off and landing equipment during the flight, and to control the UAV to land on the take-off and landing equipment based on the return trajectory.

[0078] In this application, the take-off and landing equipment is also used to send a locking command to the fastening module when it receives a prompt from the UAV indicating a successful landing. The fastening module is used to respond to the locking command and perform a locking operation on the UAV.

[0079] Specifically, after the UAV completes a soft landing and stabilizes on the surface of the landing gear, the flight controller will automatically send a landing completion signal to the landing gear. Upon receiving the landing completion signal, the landing gear sends a locking command to the fastening module. Then, the fastening module responds to the locking command by making small, precise movements in the X and Y directions to achieve precise alignment and engagement with the UAV's landing gear.

[0080] The spherical structure at the top of the fastening module helps with initial alignment with the drone's landing gear. The lower part of the fastening module has an arc-shaped structure, which provides stable support and fixation after attachment.

[0081] In addition, a mechanical spring mechanism is provided at the bottom of the fastening module. After the fastening module and the drone's landing gear are precisely aligned and engaged, the mechanical spring mechanism will automatically extend and quickly and reliably engage and lock with the drone's landing gear.

[0082] In conjunction with the above embodiments, in one implementation, the take-off and landing equipment further includes a first inertial measurement module, a first real-time dynamic positioning reference station, and a wireless transmission module; the UAV includes a second real-time dynamic positioning reference station and a second inertial measurement module, such as... Figure 2 As shown.

[0083] The first inertial measurement module is used to acquire dynamic status data of the take-off and landing equipment; The first real-time dynamic positioning reference station is used to obtain the position of the take-off and landing equipment; The wireless transmission module is used to send the dynamic status data and location of the take-off and landing equipment to the drone; The second real-time dynamic positioning base station is used to obtain the location of the UAV; The second inertial measurement module is used to acquire the dynamic state data of the UAV; The flight controller is also used to determine the return trajectory during the flight of the UAV based on the UAV's position, the UAV's dynamic status data, the position of the spherical marker, the position of the visual recognition marker, the position of the take-off and landing equipment, and the dynamic status data of the take-off and landing equipment, and to control the UAV to land on the take-off and landing equipment based on the return trajectory.

[0084] The control process of the flight controller can be referred to above.

[0085] In this application, the main body of the take-off and landing equipment is a square take-off and landing module, on which an IMU+RTK base station, wireless transmission equipment, etc., are installed, together constituting the data acquisition equipment of the take-off and landing equipment. The IMU+RTK base station is used to collect IMU and GPS data and transmit the data in real time to the UAV's flight controller for processing to calculate the current position and attitude information of the take-off and landing equipment. At the same time, the take-off and landing equipment is used to confirm its own direction of travel so that the nose of the UAV is aligned with the nose of the vehicle (i.e., heading alignment) during the UAV landing phase, ensuring the normal execution of subsequent automatic anchoring and charging operations.

[0086] A visual positioning marker is located in the center of the front of the square take-off and landing module, used for visual identification during drone descent to achieve precise landing. Movable fastening modules are installed at the four corners of the square take-off and landing module. These fastening modules can move a certain distance in the forward, backward, left, and right directions to secure the drone after landing. The top of the fastening module is designed in a spherical or similar shape (spherical marker), and the bottom is curved to fit snugly against the drone's landing gear.

[0087] During the return-to-home control of the drone, the flight controller uses the drone's downward-facing camera to identify the spherical shape on top of the fastening module. Combined with the YOLO algorithm after initial RTK positioning, it identifies the correct landable target, ensuring the reliability and robustness of landing identification and improving the anti-interference and positioning accuracy of long-range target identification. The visual positioning marker in the middle is used for high-precision real-time pose calculation at close range, meeting the high real-time requirements during landing. The mechanical spring at the bottom of the fastening module connects to the drone's landing gear, enabling contact with the positive and negative terminals of the drone's battery and the battery balance head for automatic charging. The docking direction of the positive and negative terminals is determined by the IMU of the takeoff and landing equipment to ensure correct connection.

[0088] The takeoff and landing procedures for drones are as follows: Initialization: Before takeoff or landing, initialize the drone and takeoff and landing equipment, and establish communication between the drone, takeoff and landing equipment and ground control equipment.

[0089] Takeoff Preparation and Takeoff: After initialization, the ground control equipment first sends a command to release the drone from its anchors. The takeoff and landing equipment receives the command and moves to its extreme position to prepare for subsequent landing identification. Upon receiving the release command, the drone takes off from the takeoff and landing equipment and begins its mission.

[0090] Return and Initial Identification: After completing its mission, the drone's flight controller acquires the real-time location information of the takeoff and landing equipment and returns to the area above the equipment. At this point, the drone activates its downward-facing camera and uses the YOLO algorithm to identify the spherical marker on top of the fastening module, completing the initial identification of the landing target before initiating the landing procedure.

[0091] Precise Landing: When the drone reaches a landing altitude of approximately 10 meters, the visual positioning marker positioning procedure is initiated. The flight controller acquires its position and attitude information relative to the takeoff and landing equipment in real time, and completes a precise landing on the equipment based on this information.

[0092] Automatic Securing and Charging: Once the drone has successfully landed and stabilized on the landing gear, the landing gear sends a locking command. The securing module receives the command and activates, locking the drone's landing gear and completing the automatic locking process. Simultaneously, the landing gear automatically detects the drone's battery level; if charging is required, it initiates the charging process via ground control equipment.

[0093] The technical solution of this application has at least the following technical effects: First: This application successfully integrates positioning data from RTK-GNSS, IMU, and vision processing modules, achieving high-precision fusion processing of multi-sensor data. This significantly improves the take-off and landing positioning accuracy of UAVs in dynamic platform environments, overcoming the shortcomings of traditional single or simple positioning methods.

[0094] Second: This application creatively proposes a combined recognition strategy that combines three-dimensional spherical targets (spherical markers) with two-dimensional visual positioning marks. This strategy can effectively enhance the accuracy and robustness of the visual processing module in recognizing targets at long distances, improve anti-interference capabilities, and solve the problem of poor recognition performance of traditional two-dimensional marks at long distances and large angles.

[0095] Third: This application employs a dynamic predictive control method based on extended Kalman filtering. By predicting the motion state of the vehicle platform in real time, the UAV can adjust its flight state and trajectory in advance. This predictive control mechanism effectively improves the stability of the UAV during landing and its adaptability to the mobile platform.

[0096] Fourth: This application provides a movable spring-loaded locking mechanism for automatically securing drones, which solves the problems of low automation and unreliability in the existing drone landing fixing methods, ensuring the stability of drones under vehicle movement or vibration conditions, and greatly reducing the need for manual intervention.

[0097] Fifth: The vehicle-mounted UAV flight control system provided in this application exhibits a high level of automation, capable of autonomously completing the entire process from takeoff to landing, anchoring, and even charging. With its high-precision positioning, robust recognition, and stable control capabilities in dynamic environments, it can be applied to UAV takeoff and landing scenarios in various dynamic platform environments.

[0098] Figure 3 This is a schematic diagram of the structure of a flight controller according to an embodiment of this application, as shown below. Figure 3 As shown, the flight controller may include a processor 310, a communications interface 320, a memory 330, and a communications bus 340, wherein the processor 310, the communications interface 320, and the memory 330 communicate with each other through the communications bus 340. The processor 310 can call logic instructions in the memory 330 to execute a vehicle-mounted unmanned aerial vehicle (UAV) flight control method. The method includes: when it is determined that a return-to-home operation will be performed on the UAV, determining a first return-to-home trajectory for the UAV towards the take-off and landing device based on the UAV's position and the position of the take-off and landing device adapted to the UAV, and controlling the UAV to fly according to the first return-to-home trajectory, wherein the take-off and landing device is mounted on a vehicle and the vehicle is in motion; when the UAV is flying along the first return-to-home trajectory, if a spherical marker is detected on the take-off and landing device, determining a second return-to-home trajectory for the UAV towards the take-off and landing device based on the UAV's position and the detected position of the spherical marker, and controlling the UAV to fly according to the second return-to-home trajectory; when the UAV is flying along the second return-to-home trajectory, if a visual positioning marker is detected on the take-off and landing device, controlling the UAV to land on the take-off and landing device based on the UAV's position and the detected position of the visual positioning marker.

[0099] Furthermore, the logical instructions in the aforementioned memory 330 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0100] On the other hand, this application also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute a vehicle-mounted unmanned aerial vehicle (UAV) flight control method provided by the above methods. The method includes: when it is determined that a return-to-home operation will be performed on the UAV, determining a first return-to-home trajectory for the UAV to fly towards the take-off and landing equipment based on the position of the UAV and the position of the take-off and landing equipment adapted to the UAV, and controlling the UAV to fly according to the first return-to-home trajectory, wherein the take-off and landing equipment is mounted on a vehicle and the vehicle is in motion; when the UAV is flying along the first return-to-home trajectory, if a spherical marker is detected on the take-off and landing equipment, determining a second return-to-home trajectory for the UAV to fly towards the take-off and landing equipment based on the position of the UAV and the position of the detected spherical marker, and controlling the UAV to fly according to the second return-to-home trajectory; when the UAV is flying along the second return-to-home trajectory, if a visual positioning marker is detected on the take-off and landing equipment, controlling the UAV to land on the take-off and landing equipment based on the position of the UAV and the position of the detected visual positioning marker.

[0101] Furthermore, this application also provides a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements a vehicle-mounted unmanned aerial vehicle (UAV) flight control method provided by the methods described above. This method includes: when it is determined that a return-to-home operation will be performed on the UAV, determining a first return-to-home trajectory for the UAV towards the take-off and landing equipment based on the UAV's position and the position of a take-off and landing equipment adapted to the UAV, and controlling the UAV to fly according to the first return-to-home trajectory, wherein the take-off and landing equipment is mounted on a vehicle and the vehicle is in motion; when the UAV is flying along the first return-to-home trajectory, if a spherical marker is detected on the take-off and landing equipment, determining a second return-to-home trajectory for the UAV towards the take-off and landing equipment based on the UAV's position and the detected position of the spherical marker, and controlling the UAV to fly according to the second return-to-home trajectory; when the UAV is flying along the second return-to-home trajectory, if a visual positioning marker is detected on the take-off and landing equipment, controlling the UAV to land on the take-off and landing equipment based on the UAV's position and the detected position of the visual positioning marker.

[0102] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0103] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0104] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A flight control method for a vehicle-mounted unmanned aerial vehicle, characterized in that, The method, applied to a flight controller in a drone, includes: When it is determined that a return-to-home operation will be performed on the drone, a first return-to-home trajectory for the drone to fly toward the take-off and landing equipment is determined based on the location of the drone and the location of the take-off and landing equipment adapted to the drone, and the drone is controlled to fly according to the first return-to-home trajectory. The take-off and landing equipment is mounted on a vehicle and the vehicle is in motion. During the flight of the UAV along the first return trajectory, if a spherical marker is detected on the take-off and landing equipment, a second return trajectory for the UAV to fly toward the take-off and landing equipment is determined based on the position of the UAV and the position of the detected spherical marker, and the flight of the UAV is controlled according to the second return trajectory; During the flight of the UAV along the second return trajectory, if a visual positioning marker is detected on the take-off and landing equipment, the UAV is controlled to land on the take-off and landing equipment based on the position of the UAV and the position of the detected visual positioning marker.

2. The vehicle-mounted unmanned aerial vehicle flight control method according to claim 1, characterized in that, Determining the first return trajectory of the UAV towards the take-off and landing equipment based on the UAV's location and the location of the take-off and landing equipment adapted to the UAV includes: The dynamic status data of the UAV and the dynamic status data of the take-off and landing equipment are obtained. The dynamic status data of the UAV describes the motion law of the UAV, and the dynamic status data of the take-off and landing equipment describes the motion law of the take-off and landing equipment. Based on the location of the UAV, the location of the take-off and landing equipment, the dynamic status data of the UAV, and the dynamic status data of the take-off and landing equipment, the first return trajectory is determined by an extended Kalman filter algorithm.

3. The vehicle-mounted unmanned aerial vehicle flight control method according to claim 1, characterized in that, Determining the second return trajectory of the drone towards the take-off and landing equipment based on the drone's position and the detected position of the spherical marker includes: Acquire the dynamic status data of the UAV and the dynamic status data of the take-off and landing equipment; Based on the location of the UAV, the location of the spherical marker, the dynamic status data of the UAV, and the dynamic status data of the take-off and landing equipment, the second return trajectory is determined using an extended Kalman filter algorithm.

4. The vehicle-mounted unmanned aerial vehicle flight control method according to any one of claims 1-3, characterized in that, The step of controlling the drone to land on the take-off and landing equipment based on the drone's position and the detected position of the visual positioning marker includes: With the goal of aligning the center point of the UAV with the center point of the visual positioning mark, and controlling the nose direction of the UAV to be consistent with the movement direction of the take-off and landing equipment, the actions of the UAV are adjusted according to the position of the UAV and the position of the visual positioning mark. After the adjustment is completed, the drone is controlled to land on the take-off and landing equipment.

5. The vehicle-mounted unmanned aerial vehicle flight control method according to claim 4, characterized in that, The step of adjusting the drone's movements based on its position and the position of the visual positioning mark, with the goal of aligning the drone's center point with the center point of the visual positioning mark and controlling the drone's nose direction to be consistent with the movement direction of the take-off and landing equipment, includes: Based on the position of the UAV and the position of the visual positioning marker, the lateral drift error of the UAV is determined, and the actions of the UAV are adjusted according to the lateral drift error until the lateral drift error is less than a first error threshold. The yaw angle error of the UAV is determined based on the deviation between the UAV's nose direction and the direction of movement of the take-off and landing equipment, and the UAV's actions are adjusted according to the yaw angle error until the yaw angle error is less than a second error threshold.

6. The vehicle-mounted unmanned aerial vehicle flight control method according to any one of claims 1-3, characterized in that, The control of the drone to land on the take-off and landing equipment includes: If the distance between the drone and the take-off and landing equipment is less than a preset height threshold, the descent speed of the drone is determined based on the distance between the drone and the take-off and landing equipment. Based on the descent speed, the drone is controlled to land on the take-off and landing equipment.

7. The vehicle-mounted unmanned aerial vehicle flight control method according to any one of claims 1-3, characterized in that, The control of the drone to land on the take-off and landing equipment includes: During the process of the drone landing on the take-off and landing equipment, first disturbance information caused by the take-off and landing equipment to the drone and second disturbance information caused by the environment in which the drone is located to the drone are acquired. The actions of the UAV are adjusted with the goal of eliminating the first and second disturbance information.

8. A vehicle-mounted unmanned aerial vehicle (UAV) flight control system, characterized in that, It includes ground control equipment, a drone, and take-off and landing equipment, wherein the ground control equipment is communicatively connected to the drone and the take-off and landing equipment respectively; The ground control equipment is used to control the operating status of the UAV and the take-off and landing equipment; The take-off and landing equipment has a built-in square take-off and landing module and a fastening module. The surface of the square take-off and landing module is provided with visual identification marks. The fastening modules are respectively located at the four corners of the square take-off and landing module. The top of the fastening module is provided with a spherical marker. The take-off and landing equipment is used to provide take-off and landing sites to the UAV under the control of the ground control equipment, and to send its own position to the UAV. The take-off and landing equipment is mounted on a vehicle. The drone is used to perform flight missions under the control of the ground control equipment. The drone has a built-in flight controller, which is used to determine the return trajectory based on the drone's position, the position of the spherical marker, the position of the visual recognition mark, and the position of the take-off and landing equipment during the drone's flight, and to control the drone to land on the take-off and landing equipment based on the return trajectory.

9. The vehicle-mounted unmanned aerial vehicle flight control system according to claim 8, characterized in that, The take-off and landing equipment is also equipped with a first inertial measurement module, a first real-time dynamic positioning reference station, and a wireless transmission module. The UAV also includes a second real-time dynamic positioning reference station and a second inertial measurement module. The first inertial measurement module is used to acquire the dynamic status data of the take-off and landing equipment; The first real-time dynamic positioning reference station is used to obtain the position of the take-off and landing equipment; The wireless transmission module is used to send the dynamic status data of the take-off and landing equipment and the location of the take-off and landing equipment to the UAV; The second real-time dynamic positioning reference station is used to obtain the position of the UAV; The second inertial measurement module is used to acquire the dynamic state data of the UAV; The flight controller is also used to determine the return trajectory based on the drone's position, the drone's dynamic status data, the position of the spherical marker, the position of the visual recognition mark, the position of the take-off and landing equipment, and the dynamic status data of the take-off and landing equipment during the drone's flight, and to control the drone to land on the take-off and landing equipment based on the return trajectory.

10. The vehicle-mounted unmanned aerial vehicle flight control system according to claim 8, characterized in that, The take-off and landing equipment is also used to send a locking command to the fastening module when it receives a prompt from the UAV indicating a successful landing. The fastening module is used to respond to the locking command and perform a locking operation on the UAV.

Images