Unmanned aerial vehicle air docking control method

Abstract

The invention discloses an unmanned aerial vehicle air docking control method, and the method comprises the steps: controlling the flight states of a follow-flight unmanned aerial vehicle and a pilot-flight unmanned aerial vehicle through formation flight; sending a docking command to the follow-flight unmanned aerial vehicle to control the follow-flight unmanned aerial vehicle to keep a first preset distance from the leading unmanned aerial vehicle; within the first preset distance, a camera of the follow-flight unmanned aerial vehicle is controlled to capture the taper sleeve target, after the taper sleeve target is captured, the follow-flight unmanned aerial vehicle and the leading unmanned aerial vehicle are controlled to keep a second preset distance, the air docking task is executed, and the second preset distance is smaller than the first preset distance. The method is simple in processing, higher in calculation speed and higher in output frequency, the air docking precision within 0.15 m can be achieved by combining a flight control system, and the reliability of air precise docking of the unmanned aerial vehicle is improved.

Description

Technical Field

[0001] The present invention relates to the field of aircraft technology, and more specifically, to a method for controlling aerial docking of an unmanned aerial vehicle (UAV). Background Art

[0002] Currently, aerial docking technology is primarily used for aerial refueling missions. With the development of drone technology, drones are gradually replacing manned aircraft in some covert and dangerous missions. Applying aerial docking technology to the drone sector could enable high-precision, low-risk autonomous docking missions, reducing the cost of aerial operations and improving combat capabilities.

[0003] The key and challenge of autonomous aerial docking technology for drones lies in the navigation involved in the process. This challenge lies in accurately determining the relative positions of the lead and follow drones in real time. Currently, the most widely used outdoor drone navigation and positioning technologies are satellite navigation and real-time dynamic measurement and navigation. Satellite navigation offers meter-level accuracy, while real-time dynamic measurement and navigation can achieve centimeter-level accuracy. Satellite navigation can be used for long-distance navigation, while real-time dynamic measurement and navigation can be used for closer range. However, real-time dynamic measurement and navigation have low data update rates and limited positioning accuracy, making it incapable of achieving precise docking with a 0.3m diameter drogue.

[0004] However, common visual positioning methods require establishing the positional relationship between the drone camera and the marker, and obtaining the drone's relative posture information through coordinate transformation. The visual processing algorithm involves image preprocessing, feature recognition, feature matching, coordinate transformation, and posture extraction, and the calculation process is relatively complex. Summary of the Invention

[0005] The present invention provides an image recognition-based UAV aerial docking control method to solve at least one of the problems existing in the prior art.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] The first aspect of the present invention provides a method for controlling a drone's aerial docking, the method comprising:

[0008] Control the flight status of the follower drone and the leader drone through formation flying;

[0009] Sending a docking command to the following drone to control the following drone to maintain a first preset distance from the leading drone;

[0010] Within a first preset distance, the camera of the following drone is controlled to capture the drogue target. After capturing the drogue target, the following drone is controlled to maintain a second preset distance from the leading drone to perform an aerial docking mission, wherein the second preset distance is smaller than the first preset distance.

[0011] Optionally, controlling the following drone to maintain a first preset distance from the leading drone includes controlling the following drone to move closer to the leading drone and maintain the first preset distance from the leading drone based on guidance of real-time dynamic measurement navigation information.

[0012] Optionally, the method further includes

[0013] If the camera of the following UAV fails to capture the drogue target within the first preset distance, the following UAV continues to be controlled to fly and capture according to the real-time dynamic measurement navigation information within the first preset distance.

[0014] Optionally, the docking area of ​​the docking task is a cylindrical area, and the cylindrical axis of the cylindrical area is coaxial with the cone sleeve target axis.

[0015] Optionally, the execution of the docking task includes:

[0016] Target recognition is performed based on the visual positioning image recognition algorithm to obtain the distance information of the following UAV relative to the drogue target in the horizontal and vertical directions as the first position deviation information;

[0017] Performing axial control on the following UAV according to the real-time dynamic measurement navigation information to obtain axial position information of the following UAV and the leading UAV as second position deviation information;

[0018] The axial distance between the following UAV and the leading UAV is kept fixed, and the following UAV is controlled by the first position deviation information and the second position deviation information to complete the docking task of the following UAV and the leading UAV.

[0019] Optionally, obtaining the distance information of the following UAV relative to the drogue target in the horizontal and vertical directions includes obtaining the position deviation of the following UAV relative to the drogue target in the horizontal and vertical directions according to the principle of similar triangles in camera imaging, and the formula is:

[0020]

[0021] Among them, S0 is the pixel area of ​​the circumscribed rectangular recognition box of the visual positioning image recognition algorithm in the camera imaging plane, d is the actual diameter of the cone target, u1 is the horizontal pixel of the cone target center distance image, v1 is the vertical pixel of the cone target center distance image, x1 is the horizontal position deviation of the following UAV relative to the cone target, and y1 is the vertical position deviation of the following UAV relative to the cone target.

[0022] Optionally, completing the docking task of the follower drone and the lead drone further includes determining whether the docking accuracy and docking time requirements are met.

[0023] If the docking accuracy and docking time requirements are met, the docking task is terminated;

[0024] If the docking accuracy and docking time requirements are not met, the first deviation information is updated based on the visual positioning image recognition algorithm, and the second deviation information is updated through real-time dynamic measurement of navigation information.

[0025] Optionally, the sending of a docking command to the following drone to control the following drone to maintain a first preset distance from the leading drone includes:

[0026] If the following UAV does not maintain the first preset distance with the leading UAV after receiving the docking command, the real-time dynamic measurement navigation information is updated to continue controlling the flight of the following UAV.

[0027] Optionally, the method further includes, after completing the docking, the following drone and the leading drone entering a separation state.

[0028] Optionally, the following drone and the leading drone entering a separation state include:

[0029] The following drone switches to real-time dynamic measurement and navigation mode, and automatically separates after the waiting time reaches the preset time, leaving the second preset distance.

[0030] The beneficial effects of the present invention are as follows:

[0031] The present invention controls the relative positions of the following UAV and the leading UAV based on a visual positioning image recognition algorithm and real-time dynamic measurement navigation information. The method has simple processing, faster calculation speed, and higher output frequency. Combined with the flight control system, it can achieve an aerial docking accuracy of less than 0.15m, thereby improving the reliability of precise aerial docking of UAVs. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] The specific embodiments of the present invention will be further described in detail below with reference to the accompanying drawings.

[0033] Figure 1 A flowchart illustrating a UAV aerial docking task according to an embodiment of the present invention is shown;

[0034] Figure 2 A flowchart illustrating a method for controlling a drone through real-time dynamic measurement and navigation in accordance with an embodiment of the present invention is shown;

[0035] Figure 3 A schematic diagram showing the positions of the precise docking area and the preset docking area in an embodiment of the present invention;

[0036] Figure 4 A schematic diagram showing the pinhole imaging principle of an airborne camera in an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0037] In order to more clearly illustrate the present invention, the present invention is further described below in conjunction with preferred embodiments and accompanying drawings. Similar components in the accompanying drawings are represented by the same reference numerals. It should be understood by those skilled in the art that the following detailed description is illustrative rather than restrictive and should not be used to limit the scope of protection of the present invention.

[0038] One embodiment of the present invention proposes a method for controlling aerial docking of a UAV, which performs target recognition through an image processing algorithm, calculates the relative position data of the UAV in the horizontal and vertical directions relative to a marker based on the principle of similar triangles of camera pinhole imaging, and provides axial data information for precise aerial docking of the UAV in combination with the axial relative data of the UAV relative to the marker provided by real-time dynamic measurement and navigation, and completes precise docking of the following UAV and the leading UAV based on the relative position data of the UAV in the horizontal and vertical directions relative to the marker and the axial relative data relative to the marker, providing technical verification for aerial operations of UAVs. GPS (Global Positioning System) is the abbreviation of Global Positioning System. The working principle of GPS is to measure the distance between a satellite with a known position and a user receiver, and then the specific position of the receiver can be known by integrating the data of multiple satellites. RTK (Real-time kinematic) is the abbreviation of real-time dynamic measurement and navigation technology. RTK is a technology that uses GPS carrier phase observations for real-time dynamic relative positioning.

[0039] In this embodiment, each drone is equipped with a camera at its head, with its field of view oriented in the direction of the drone's axial flight. The camera is used to acquire horizontal and vertical positional data relative to a marker carried by the forward-facing drone. A drogue is also installed at the drone's tail, connected to the drone via a connecting rod. The connecting rod is mounted at the tail of the drone and extends rearward along the drone's axis. The drogue is conical, with its top connected to the connecting rod and its bottom extending rearward along the drone's axis. The circular bottom serves as the drogue target, serving as a capture target during docking with other drones, facilitating the successful completion of aerial docking missions.

[0040] The method provided in this embodiment includes:

[0041] Control the flight status of the follower drone and the leader drone through formation flying;

[0042] Send a docking command to the following drone, and control the following drone to move closer to the leading drone based on the RTK navigation information, so that the following drone and the leading drone maintain a first preset distance;

[0043] If the following drone does not maintain the first preset distance from the leading drone after receiving the docking command, the RTK navigation information is updated to continue guiding the following drone.

[0044] If the camera of the following drone captures the drogue target within the first preset distance, the following drone is controlled to maintain a second preset distance from the leading drone to perform the aerial docking mission, wherein the second preset distance is smaller than the first preset distance, and the second preset distance is the docking distance between the following drone and the leading drone.

[0045] If the camera of the following UAV fails to capture the drogue target within the first preset distance, the following UAV will continue to be guided to fly according to the RTK navigation information within the first preset distance and continue to capture until the onboard camera of the following UAV captures the drogue target.

[0046] Specifically, the docking area of ​​the docking task is a cylindrical area, and the cylindrical axis of the cylindrical area is coaxial with the axis of the cone sleeve target.

[0047] When performing the docking mission, target recognition is performed based on the visual positioning image recognition algorithm to obtain the distance information of the following UAV relative to the drogue target in the horizontal and vertical directions as the first position deviation information;

[0048] Performing axial control on the following UAV according to the RTK navigation information to obtain axial position information of the following UAV and the leading UAV as second position deviation information;

[0049] The axial distance between the following UAV and the leading UAV is kept fixed, and the following UAV is controlled by the first position deviation information and the second position deviation information to complete the docking task of the following UAV and the leading UAV.

[0050] The method of obtaining the distance information of the following UAV relative to the drogue target in the horizontal and vertical directions based on the visual positioning image recognition algorithm includes obtaining the position deviation of the following UAV relative to the drogue target in the horizontal and vertical directions according to the principle of similar triangles in camera imaging. The formula is:

[0051]

[0052] Among them, S0 is the pixel area of ​​the circumscribed rectangular recognition box of the visual positioning image recognition algorithm in the camera imaging plane, d is the actual diameter of the cone target, u1 is the horizontal pixel of the cone target center distance image, v1 is the vertical pixel of the cone target center distance image, x1 is the horizontal position deviation of the following UAV relative to the cone target, and y1 is the vertical position deviation of the following UAV relative to the cone target.

[0053] When completing the docking mission between the follower drone and the leader drone, it is also necessary to determine whether the docking accuracy and docking time requirements are met:

[0054] If the docking accuracy and docking time requirements are met, the docking task is terminated;

[0055] If the docking accuracy and docking time requirements are not met, the first deviation information is updated based on the visual positioning image recognition algorithm, and the second deviation information is updated using the RTK navigation information.

[0056] After completing the precise docking, the following drone and the leading drone enter the separation state. The following drone switches to RTK navigation mode and automatically separates after the waiting time reaches the preset time. The following drone separates at the second preset distance and returns or performs other tasks.

[0057] In a specific embodiment, based on the high precision and high frequency characteristics of the visual algorithm, a visual sensor and an RTK system are used to obtain the relative position of the following drone and the docking drogue, providing reliable navigation information for the aerial docking control system. The docking task of this embodiment is to achieve a docking accuracy of less than 0.15m within a certain period of time for a drogue target with a diameter of 0.3m. The method of performing the aerial docking task of the drone is divided into three stages, such as Figure 1 As shown, specifically:

[0058] 1. Control the flight status of the follower drone and the leader drone through formation flying

[0059] The drone formation flight phase refers to the formation flight process in which the leading drone and the following drone maintain a certain formation and distance before the following drone receives the docking command. This phase is to ensure that the leading drone and the following drone maintain normal flight status.

[0060] 2. Control the follower drone to maintain the first preset distance from the lead drone

[0061] A docking command is sent to the following drone. When the following drone receives the docking command during the formation flight phase, it is controlled to move closer to the leading drone based on the guidance of the RTK navigation information, so that the following drone maintains a first preset distance from the leading drone, that is, the following drone enters the preset docking area.

[0062] 3. Control the follower drone to maintain a second preset distance from the lead drone and perform the aerial docking mission

[0063] In the preset docking area, determine whether the drone's onboard camera captures the drogue target:

[0064] If the follower drone's onboard camera fails to capture the drogue target within the preset docking area, the follower drone will continue to fly and capture the target based on the RTK navigation information within the preset docking area.

[0065] If the onboard camera of the follow-flying UAV captures the drogue target within the preset distance docking area, the follow-flying UAV will be controlled to maintain a second preset distance with the leading UAV, that is, the precise docking area, to perform the aerial docking mission.

[0066] In this embodiment, the precise docking area is a cylindrical area behind the drogue target, with the cylinder axis coaxially arranged with the drogue target axis. The drogue target has a diameter of 0.3m, and the cylinder has a diameter of 0.5m. During the precise docking phase, the horizontal and vertical positional deviations of the following drone relative to the drogue target are calculated based on the results of an image processing algorithm. This is used as first positional deviation information, and the following drone is controlled laterally and longitudinally using this first positional deviation information. Based on RTK navigation information, the following drone is provided with an axial positional deviation from the leading drone as second positional deviation information. This second positional deviation information is used to control the following drone's axial flight, causing it to move closer to the leading drone. The first and second positional deviation information are used to jointly control the following drone's docking flight.

[0067] Figure 2This is a framework diagram of the drone docking control process. According to the RTK navigation, the flight of the follow-flying drone is controlled, and it is judged whether the follow-flying drone and the leading drone maintain the first preset distance: if the follow-flying drone and the leading drone do not maintain the first preset distance, the flight of the follow-flying drone is continued to be controlled according to the RTK navigation; if the follow-flying drone and the leading drone maintain the first preset distance, it is judged whether the airborne camera of the follow-flying drone captures the drogue target within the first preset distance: if the camera of the follow-flying drone does not capture the drogue target, the flight of the follow-flying drone is continued to be controlled according to the RTK navigation information and continues to capture it within the first preset distance; if the camera of the follow-flying drone captures the drogue target, the follow-flying drone is The UAV maintains a second preset distance from the leading UAV and starts to perform the docking task; when performing the docking task, target recognition is performed according to the image recognition algorithm in visual positioning, and the position deviation information of the following UAV relative to the drogue target in the horizontal and vertical directions is obtained, and the axial flight of the following UAV is controlled according to the RTK navigation to obtain the axial position deviation information of the following UAV; the following UAV and the leading UAV are kept at a second preset distance, and it is determined whether the docking accuracy and docking time requirements are met. In this embodiment, the diameter of the drogue target is 0.3m, and the docking accuracy is achieved within 0.15m within the preset time, which means that the docking requirements are met and the docking task is ended.

[0068] Figure 3 A schematic diagram showing the positions of the precise docking area and the preset docking area in an embodiment of the present invention is shown, wherein the following drone maintains a first preset distance from the leading drone, that is, the following drone enters the preset docking area, and the following drone maintains a second preset distance from the leading drone, that is, the following drone enters the precise docking area.

[0069] During the docking mission phase, the visual positioning algorithm serves as the input of the control system. The image processing algorithm in the visual positioning algorithm provides the lateral and longitudinal distance information of the following UAV from the drogue target. The visual positioning logic is divided into ground reference pixel information calibration and airborne real-time visual positioning.

[0070] In the ground reference pixel information calibration stage, according to Figure 4 As shown in the schematic diagram of pinhole imaging of the airborne camera, assume that the focal length of the following drone camera is f, the distance between the camera and the cone target is D, the actual diameter of the cone target is d, the area of ​​its circumscribed rectangle is S, the pixel area of ​​the circumscribed rectangle recognition box of the image recognition algorithm in the camera imaging plane is S0, the horizontal pixel distance from the center of the cone target in the image is u0, and the vertical pixel distance from the center of the cone target in the image is v0.

[0071] In the real-time visual positioning phase on the airborne side, let the distance between the following drone camera and the drogue target be D1, the horizontal position deviation of the following drone relative to the drogue target be x1, the vertical position deviation of the following drone relative to the drogue target be y1, the pixel area of ​​the circumscribed rectangular identification box of the drogue target in the pixel plane be S1, the horizontal pixel distance from the center of the drogue target to the image be u1, and the vertical pixel distance from the center of the drogue target to the image be v1. According to the principle of similar triangles in camera imaging, we can obtain:

[0072]

[0073] The position deviation of the following UAV relative to the drogue target in the horizontal and vertical directions is calculated, and then the following UAV is controlled in the lateral and longitudinal directions.

[0074] The axial distance between the following UAV and the leading UAV is kept fixed, and the docking flight of the following UAV and the leading UAV is controlled by the first position deviation information and the second position deviation information to achieve a precise docking task.

[0075] When the precise docking requirements are met, the separation phase begins. The following drone switches to RTK navigation mode, leaves the precise docking area, and returns home or performs other tasks.

[0076] The following drone can be separated automatically during the separation phase, or it can be separated after receiving a control signal. In this embodiment, when the docking accuracy is met, the following drone will automatically separate after the waiting time reaches a preset time.

[0077] The method provided in this embodiment can control the relative positions of the following drone and the leading drone based on the visual positioning image recognition algorithm and RTK navigation information. This method is simple to process, has a faster calculation speed, and a higher output frequency, which improves the reliability of precise aerial docking of drones and provides technical verification for drone aerial operations. The method provided in this embodiment can be applied to a series of fields such as aerial refueling, refueling, capture, and recovery of drones.

[0078] In the description of the present invention, it should be noted that the orientation or positional relationship indicated by the terms "upper" and "lower" is based on the orientation or positional relationship shown in the accompanying drawings, and is only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention. Unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be internal communication between two elements. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to the specific circumstances.

[0079] It should also be noted that, in the description of the present invention, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or elements inherent to such process, method, article, or device. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the process, method, article, or device comprising the element.

[0080] Obviously, the above embodiments of the present invention are merely examples for clearly illustrating the present invention, and are not limitations on the implementation methods of the present invention. For ordinary technicians in the relevant field, other different forms of changes or modifications can be made based on the above description. It is impossible to list all the implementation methods here. All obvious changes or modifications derived from the technical solution of the present invention are still within the scope of protection of the present invention.

Claims

1. A method for controlling aerial docking of a UAV, characterized in that: The method includes Control the flight status of the follower drone and the leader drone through formation flying; Sending a docking command to the following drone to control the following drone to maintain a first preset distance from the leading drone; Within a first preset distance, the camera of the following drone is controlled to capture the drogue target. After capturing the drogue target, the following drone is controlled to maintain a second preset distance from the leading drone to perform an aerial docking mission, wherein the second preset distance is smaller than the first preset distance.

2. The method according to claim 1, characterized in that Controlling the following drone to maintain the first preset distance from the leading drone includes controlling the following drone to move closer to the leading drone and maintain the first preset distance from the leading drone based on the guidance of the real-time dynamic measurement navigation information.

3. The method according to claim 1, characterized in that The method further includes If the camera of the following UAV fails to capture the drogue target within the first preset distance, the following UAV continues to be controlled to fly and capture according to the real-time dynamic measurement navigation information within the first preset distance.

4. The method according to claim 1, wherein The docking area of ​​the docking task is a cylindrical area, and the cylindrical axis of the cylindrical area is coaxial with the cone sleeve target axis.

5. The method according to claim 1, characterized in that The execution of the docking task includes Target recognition is performed based on the visual positioning image recognition algorithm to obtain the distance information of the following UAV relative to the drogue target in the horizontal and vertical directions as the first position deviation information; Performing axial control on the following UAV according to the real-time dynamic measurement navigation information to obtain axial position information of the following UAV and the leading UAV as second position deviation information; The axial distance between the following UAV and the leading UAV is kept fixed, and the following UAV is controlled by the first position deviation information and the second position deviation information to complete the docking task of the following UAV and the leading UAV.

6. The method according to claim 5, characterized in that The obtaining of the distance information of the following UAV relative to the drogue target in the horizontal and vertical directions includes obtaining the position deviation of the following UAV relative to the drogue target in the horizontal and vertical directions according to the principle of similar triangles in camera imaging, and the formula is: Among them, S0 is the pixel area of ​​the circumscribed rectangular recognition box of the visual positioning image recognition algorithm in the camera imaging plane, d is the actual diameter of the cone target, u1 is the horizontal pixel of the cone target center distance image, v1 is the vertical pixel of the cone target center distance image, x1 is the horizontal position deviation of the following UAV relative to the cone target, and y1 is the vertical position deviation of the following UAV relative to the cone target.

7. The method according to claim 5, characterized in that The task of completing the docking of the follow-flying UAV and the leading UAV also includes determining whether the docking accuracy and docking time requirements are met. If the docking accuracy and docking time requirements are met, the docking task is terminated; If the docking accuracy and docking time requirements are not met, the first deviation information is updated based on the visual positioning image recognition algorithm, and the second deviation information is updated through real-time dynamic measurement of navigation information.

8. The method according to claim 1, characterized in that The sending of a docking command to the following drone to control the following drone to maintain a first preset distance from the leading drone includes: If the following UAV does not maintain the first preset distance with the leading UAV after receiving the docking command, the real-time dynamic measurement navigation information is updated to continue controlling the flight of the following UAV.

9. The method according to claim 1, characterized in that The method also includes, after the docking is completed, the following drone and the leading drone entering a separation state.

10. The method according to claim 9, characterized in that The following UAV and the leading UAV entering a separation state include: The following drone switches to real-time dynamic measurement and navigation mode, and automatically separates after the waiting time reaches the preset time, leaving the second preset distance.

Images