A visual information target tracking guidance method for anti-drone systems

Abstract

The application discloses a visual information target tracking guidance method for an anti-UAV system, and belongs to the field of UAV guidance control. The method obtains target two-dimensional image information through a monocular camera, obtains boundary box pixel coordinates through target detection and converts the boundary box pixel coordinates into normalized image coordinates, and estimates relative distance in combination with target real area and camera parameters; image servo control IBVS is adopted to extract geometric guidance information, centripetal acceleration is calculated through fusion of nonlinear path tracking guidance NPFG; based on the coordinated turning idea, horizontal and lateral control instructions are generated, longitudinal and vertical control instructions are generated at the same time, and finally expected roll angle, yaw angle rate, forward speed and vertical speed are output to realize closed-loop guidance. The application does not need three-dimensional position and depth information, has low computing power consumption and strong real-time performance, can eliminate high-speed flight sideslip and attitude oscillation, realizes stable tracking of a high-speed UAV target in a windy environment, reduces system hardware cost, and improves anti-interference and maneuvering tracking capability.

Description

Technical Field

[0001] This invention relates to the field of unmanned aerial vehicle (UAV) guidance and control technology, and in particular to a visual information target tracking and guidance method for anti-UAV systems. Background Technology

[0002] With the widespread application of low-altitude drones, the demand for high-speed drone guidance and tracking in counter-drone systems is increasing. Existing target tracking and guidance methods mainly fall into two categories: one is position-based image servoing and trajectory planning methods, which rely on 3D reconstruction and target depth information, resulting in high computational consumption, high control latency, susceptibility to oscillation and divergence, and high hardware costs; the other is pure image-visual servoing (IBVS) methods, which are prone to system failure after target loss, and high-speed lateral tracking easily leads to sideslip, causing a sharp increase in aerodynamic drag, rotor wake interference, thrust saturation, and even loss of control and crash. Furthermore, traditional methods struggle to achieve smooth and stable guidance in high-speed flight and windy environments, failing to meet the high-dynamic, high-real-time, and low-cost tracking requirements of counter-drone systems. Summary of the Invention

[0003] This invention provides a visual information target tracking and guidance method for anti-drone systems that can achieve smooth and stable target tracking by high-speed UAVs in windy environments by relying solely on monocular two-dimensional visual information without the need for depth perception.

[0004] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0005] This invention provides a visual information target tracking and guidance method and system for anti-drone systems, comprising:

[0006] S1: Obtain the target bounding box pixel information in the image through the target detection algorithm, calculate the center point pixel coordinates and convert them into normalized image coordinates, and estimate the physical distance from the UAV to the target by combining the actual area of ​​the target and the camera intrinsic parameters.

[0007] S2: Based on image servo control IBVS, visual features are extracted and combined with nonlinear path tracking guidance NPFG to calculate the centripetal acceleration required to cut into the target trajectory;

[0008] S3: Based on coordinated turning mechanics, the desired roll angle and desired yaw rate are generated to form lateral control commands and eliminate sideslip during high-speed flight of the UAV;

[0009] S4: The controller generates the desired forward velocity and desired vertical velocity, forming longitudinal and vertical control commands;

[0010] S5: Send the desired roll angle, desired yaw rate, desired forward velocity, and desired vertical velocity to the underlying controller to complete target tracking and guidance.

[0011] Furthermore, the physical distance mentioned in step S1 is calculated jointly by the target bounding box pixel area, the target's actual physical area, and the camera's equivalent focal length.

[0012] Furthermore, the centripetal acceleration mentioned in step S2 is calculated from the UAV's horizontal ground speed, forward-looking distance, and the sine value of the line-of-sight angle.

[0013] Furthermore, the desired roll angle mentioned in step S3 is calculated from the centripetal acceleration and gravitational acceleration, and the desired yaw rate is used to keep the camera aligned with the target.

[0014] Furthermore, in step S4, a PD controller is used to generate the desired forward velocity, and a PID controller is used to generate the desired vertical velocity.

[0015] Compared with the prior art, the technical solution disclosed in this invention has the following beneficial effects:

[0016] This invention uses only a monocular camera to acquire two-dimensional image information and combines it with the actual area of ​​the target to estimate the distance. It does not require LiDAR, depth camera and three-dimensional trajectory planning, thus reducing computing power consumption and hardware costs and improving system real-time performance.

[0017] This invention integrates IBVS and NPFG guidance to calculate centripetal acceleration, avoiding sudden command changes and violent attitude oscillations in traditional methods, ensuring smooth high-speed tracking trajectory and improving tracking stability;

[0018] This invention employs coordinated turning to generate lateral commands, eliminating sideslip during high-speed flight, reducing aerodynamic drag and wake interference, and improving anti-interference capability and power efficiency in windy environments.

[0019] This invention generates complete guidance commands through hierarchical control, resulting in low target loss risk and strong control robustness, making it suitable for high-speed dynamic tracking scenarios in anti-drone systems. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a schematic flowchart of a visual information target tracking and guidance method for an anti-drone system provided in an embodiment of the present invention;

[0022] Figure 2 The flowchart illustrates the principle of a visual information target tracking and guidance method for anti-drone systems provided in this embodiment of the invention. Detailed Implementation

[0023] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0024] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0025] This invention provides a visual information target tracking and guidance method for anti-drone systems that can achieve smooth and stable target tracking by high-speed UAVs in windy environments by relying solely on monocular two-dimensional visual information without the need for depth perception.

[0026] like Figures 1-2 As shown, this embodiment of the invention provides a visual information target tracking and guidance method for anti-drone systems, comprising:

[0027] S1: Obtain the target bounding box pixel information in the image through the target detection algorithm, calculate the center point pixel coordinates and convert them into normalized image coordinates, and estimate the physical distance from the UAV to the target by combining the actual area of ​​the target and the camera intrinsic parameters.

[0028] The physical distance mentioned in step S1 is calculated by combining the pixel area of ​​the target bounding box, the actual physical area of ​​the target, and the equivalent focal length of the camera.

[0029] Specifically, using an object detection algorithm, the pixel information of the output target bounding box in the current image frame is obtained, with the top-left corner coordinates being... The width and height of the target bounding box are respectively , And calculate the center point pixel and pixel coordinates. :

[0030] Camera internal parameters as the main point Horizontal focal length vertical focal length The pixel coordinates mentioned above can be converted to normalized image coordinates. :

[0031]

[0032] Assume the actual area of ​​the target in reality is .

[0033] Calculate the pixel area of ​​the bounding box Estimate the current physical distance between the drone and the target:

[0034]

[0035] in, This is the average physical equivalent focal length of the camera.

[0036] S2: Based on image servo control IBVS, visual features are extracted and combined with nonlinear path tracking guidance NPFG to calculate the centripetal acceleration required to cut into the target trajectory;

[0037] The centripetal acceleration mentioned in step S2 is calculated from the UAV's horizontal ground speed, forward-looking distance, and the sine value of the line-of-sight angle.

[0038] Specifically, the features extracted by the IBVS algorithm are combined with NPFG to achieve UAV guidance.

[0039] Get the current horizontal ground speed of the drone .

[0040] Forward sight distance: .

[0041] Let the vertical depth of the target from the camera be . The longitudinal distance of the target deviating from the optical axis is Let the line of sight be . Then the tangent value of the line of sight angle .

[0042] From the x-coordinate of the normalized image coordinates .

[0043] Therefore there is

[0044] By basic trigonometric identities, we have the sine of the line-of-sight angle:

[0045]

[0046] in, This is the abscissa vector of the normalized image coordinates.

[0047] Therefore, the centripetal acceleration required to pull the target back to the center of the image, i.e., to cut into the target's trajectory, is:

[0048]

[0049] S3: Based on coordinated turning mechanics, the desired roll angle and desired yaw rate are generated to form lateral control commands and eliminate sideslip during high-speed flight of the UAV;

[0050] The desired roll angle mentioned in step S3 is calculated from the centripetal acceleration and the gravitational acceleration, and the desired yaw rate is used to keep the camera aligned with the target.

[0051] Specifically, lateral control commands are generated using the concept of coordinated turning to prevent the drone from sideslipping. This utilizes gravitational acceleration. In conjunction with coordinating cornering mechanics, generate the desired roll angle:

[0052]

[0053] To eliminate sideslip and keep the camera focused on the target, the desired yaw rate is calculated. :

[0054]

[0055] S4: The controller generates the desired forward velocity and desired vertical velocity, forming longitudinal and vertical control commands;

[0056] Specifically, longitudinal control commands are generated simultaneously with lateral control commands. The controller is used to calculate the desired forward velocity. For example, a PID controller or image servo control method can be used, and the desired pitch angle can be converted through the underlying flight control system. .

[0057] When the target deviates vertically from the frame, control the drone's ascent and descent, and use the controller to calculate the desired vertical velocity. .

[0058] S5: Send the desired roll angle, desired yaw rate, desired forward velocity, and desired vertical velocity to the underlying controller to complete target tracking and guidance.

[0059] Specifically, the expected roll angle of the complete set of control commands obtained Desired yaw rate Expected forward velocity Desired vertical velocity The control commands are then sent to the underlying controller for execution.

[0060] More specific examples are as follows:

[0061] This embodiment takes a certain type of UAV in a high-speed flight scenario as an example to carry out target tracking and guidance control based on two-dimensional visual information.

[0062] S1 can be defined as follows: using the YOLO object detection algorithm, the pixel information of the output target bounding box in the current image frame is obtained, and the coordinates of the top left corner are... The width and height of the target bounding box are respectively , And calculate the center point pixel and pixel coordinates. :

[0063]

[0064] Camera internal parameters as the main point Horizontal focal length vertical focal length The pixel coordinates mentioned above can be converted to normalized image coordinates. :

[0065]

[0066] Assume the actual area of ​​the target in reality is .

[0067] Calculate the pixel area of ​​the bounding box Estimate the current physical distance between the drone and the target:

[0068]

[0069] in, This is the average physical equivalent focal length of the camera.

[0070] S2 can be used to achieve UAV guidance by utilizing features extracted by the IBVS algorithm and combining them with NPFG.

[0071] Get the current horizontal ground speed of the drone .

[0072] Forward sight distance: .

[0073] Let the vertical depth of the target from the camera be . The longitudinal distance of the target deviating from the optical axis is Let the line of sight be . Then the tangent value of the line of sight angle .

[0074] From the x-coordinate of the normalized image coordinates .

[0075] Therefore there is

[0076] By basic trigonometric identities, we have the sine of the line-of-sight angle:

[0077]

[0078] in, This is the abscissa vector of the normalized image coordinates.

[0079] Therefore, the centripetal acceleration required to pull the target back to the center of the image, i.e., to cut into the target's trajectory, is:

[0080]

[0081] S3 can generate lateral control commands using the concept of coordinated turning to prevent the drone from sideslipping. It utilizes gravitational acceleration. In conjunction with coordinating cornering mechanics, generate the desired roll angle:

[0082]

[0083] To eliminate sideslip and keep the camera focused on the target, the desired yaw rate is calculated. :

[0084]

[0085] S4 can generate longitudinal control commands simultaneously with lateral control commands. Here, the controller is used to generate the desired forward horizontal ground speed. :

[0086]

[0087] in, To achieve the desired forward flight speed, Forward proportional gain, To estimate the target depth, To the desired tracking distance, Forward differential gain, The rate of change of depth.

[0088] The desired vertical velocity is generated using the controller. :

[0089]

[0090] in, For the desired vertical velocity, The vertical coordinates of the normalized image. For vertical scaling gain, For vertical integral gain, For vertical differential gain, This represents the rate of change of the longitudinal error.

[0091] S5 can be the expected roll angle of the complete set of control commands obtained. Desired yaw rate Expected forward velocity Desired vertical velocity The control commands are then sent to the underlying controller for execution. The control commands are as follows.

[0092]

[0093] Compared with the prior art, the beneficial effects of the embodiments of the present invention are as follows:

[0094] At the technical level, this invention eliminates the delays caused by the calculation of target 3D coordinates and complex trajectory planning algorithms in traditional methods, greatly improving the real-time performance and command tracking accuracy of high-speed UAVs when tracking targets; it effectively solves the problem of command explosion in traditional IBVS control when facing sudden changes, and the system can avoid drastic attitude changes without complex manual parameter tuning, ensuring smooth high-speed pursuit trajectory and safe flight; at the same time, this method ensures that the target is always in the center of the camera's field of view, improving the continuity of visual perception and the success rate of tracking.

[0095] In terms of economy and energy efficiency, this method does not require expensive depth cameras, lidar, or high-performance onboard computers. It can achieve high-speed tracking with only a monocular camera and basic computing power, which significantly saves computing energy consumption and can extend the flight time of UAVs. It eliminates the sideslip angle generated during high-speed UAV tracking, greatly reduces the interference of nonlinear aerodynamic drag on the rotor wake, improves the anti-interference ability of the aircraft in strong wind environment, and greatly improves the power utilization rate during high-speed maneuvers.

[0096] Furthermore, traditional high-speed drone tracking systems require cumbersome PID parameter adjustments. This invention, based on physical and kinematic geometric relationships, possesses built-in dynamic adaptive capabilities. This invention solves the problems of sideslip and loss of field of view in high-speed rotorcraft drones, enabling ordinary rotorcraft drones to perform tasks previously only achievable with drones equipped with expensive equipment.

[0097] The basic principles of the present invention have been described above with reference to specific embodiments. However, it should be noted that the advantages, benefits, and effects mentioned in the present invention are merely examples and not limitations, and should not be considered as essential features of each embodiment of the present invention. Furthermore, the specific details disclosed above are for illustrative and facilitative purposes only, and are not limitations. These details do not limit the present invention to the necessity of employing the aforementioned specific details.

[0098] The block diagrams of devices, apparatuses, devices, and systems involved in this invention are merely illustrative examples and are not intended to require or imply that they must be connected, arranged, or configured in the manner shown in the block diagrams. As those skilled in the art will recognize, these devices, apparatuses, devices, and systems can be connected, arranged, and configured in any manner. Words such as “comprising,” “including,” “having,” etc., are open-ended terms meaning “including but not limited to,” and are used interchangeably with them. The terms “or” and “and” as used herein refer to the terms “and / or,” and are used interchangeably with them unless the context clearly indicates otherwise. The term “such as” as used herein refers to the phrase “such as but not limited to,” and is used interchangeably with it.

[0099] It should also be noted that in the apparatus, device, and method of the present invention, the components or steps can be disassembled and / or recombined. These disassemblies and / or recombinations should be considered as equivalent solutions of the present invention.

[0100] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use the invention. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein can be applied to other aspects without departing from the scope of the invention. Therefore, the invention is not intended to be limited to the aspects shown herein, but rather to be carried out within the widest scope consistent with the principles and novel features disclosed herein.

[0101] It should be understood that the qualifying terms "first", "second", "third", "fourth", "fifth" and "sixth" used in the description of the embodiments of the present invention are only used to more clearly illustrate the technical solutions and are not intended to limit the scope of protection of the present invention.

[0102] The above description has been given for purposes of illustration and description. Furthermore, this description is not intended to limit the embodiments of the invention to the forms disclosed herein. Although numerous exemplary aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, alterations, additions, and sub-combinations therein.

Claims

1. A visual information target tracking and guidance method for anti-UAV systems, characterized in that, include: S1: Obtain the target bounding box pixel information in the image through the target detection algorithm, calculate the center point pixel coordinates and convert them into normalized image coordinates, and estimate the physical distance from the UAV to the target by combining the actual area of ​​the target and the camera intrinsic parameters. S2: Based on image servo control IBVS, visual features are extracted and combined with nonlinear path tracking guidance NPFG to calculate the centripetal acceleration required to cut into the target trajectory; S3: Based on coordinated turning mechanics, the desired roll angle and desired yaw rate are generated to form lateral control commands and eliminate sideslip during high-speed flight of the UAV; S4: The controller generates the desired forward velocity and desired vertical velocity, forming longitudinal and vertical control commands; S5: Send the desired roll angle, desired yaw rate, desired forward velocity, and desired vertical velocity to the underlying controller to complete target tracking and guidance.

2. The guidance method according to claim 1, characterized in that, The physical distance mentioned in step S1 is calculated by combining the pixel area of ​​the target bounding box, the actual physical area of ​​the target, and the equivalent focal length of the camera.

3. The guidance method according to claim 1, characterized in that, The centripetal acceleration mentioned in step S2 is calculated from the UAV's horizontal ground speed, forward-looking distance, and the sine value of the line-of-sight angle.

4. The guidance method according to claim 1, characterized in that, The desired roll angle mentioned in step S3 is calculated from the centripetal acceleration and the gravitational acceleration, and the desired yaw rate is used to keep the camera aligned with the target.

Images