WO2018014420A1

WO2018014420A1 - Light-emitting target recognition-based unmanned aerial vehicle tracking control system and method

Info

Publication number: WO2018014420A1
Application number: PCT/CN2016/097249
Authority: WO
Inventors: 王军
Original assignee: 深圳曼塔智能科技有限公司
Priority date: 2016-07-21
Filing date: 2016-08-30
Publication date: 2018-01-25
Also published as: CN106598075A

Abstract

An unmanned aerial vehicle (20) tracking control system and method based on recognition of a light-emitting target (10), the system comprising a light-emitting target (10) and an unmanned aerial vehicle (20), the unmanned aerial vehicle (20) comprising: at least two cameras (21, 22), the two cameras (21, 22) being used to image the light-emitting target (10) in real time; an operation processing unit (23) connected to the cameras (21, 22), which recognizes the light-emitting target (10) in images shot by the cameras (21, 22) and calculates motion information of the light-emitting target (10); a flight control unit (24) connected to the operation processing unit (23), which controls the unmanned aerial vehicle (20) to fly following the light-emitting target (10) according to the motion information of the light-emitting target (10). The present system solves well the problems of instability, complicated systems, high cost and low accuracy, and in addition to relatively accurate tracking, the system can also be a completely new control method, so that an unmanned aerial vehicle (20) only needs an added device to realize multiple types of smart functionality, making the system very suitable for popularizing aerial photography entertainment unmanned aerial vehicles.

Description

UAV tracking control system and method based on illuminating target recognition

[0001] The present invention relates to drone technology, and more particularly to a drone tracking control system and method based on illuminating target recognition.

Background technique

[0002] The automatic tracking function is a very popular and popular feature in drones in the past two years, especially for entertainment aerial drones. With this function, the drone becomes more intelligent, it can Automatically tracking the specified targets, realizing the functions of shooting, interaction, etc., is an indispensable intelligent function of entertainment drones.

[0003] There are two main ways to implement this function. The first one is based on GPS positioning technology. The target object to be tracked needs to carry a communication module with GPS and communication function with the drone, such as GPS-free. The human-machine remote control, the smart phone, the smart bracelet with GPS, etc., the communication module transmits the G PS information of the location of the target to the drone, and the drone relies on the location information to implement the tracking function, and the advantage is that the technology Mature and simple, the price is moderate.

[0004] The second method is a target tracking technology based on binocular visual recognition. The core of the technology is to capture the tracking target through a dual camera on the drone, and then determine the target and the drone through a complex algorithm. The previous distance, as well as the amount and direction of the target's movement, the drone control system obtains this information and then adjusts the flight direction and flight speed to achieve the tracking function. The advantage of this technology is that it is practical both indoors and outdoors. It does not require any equipment to track the target, and it can also be used to avoid obstacles.

[0005] The above two existing technologies have their own advantages and have many shortcomings. The disadvantage of the GPS tracking method is that it can only be used outdoors in an open GPS signal, which is susceptible to environmental interference, and is subject to environmental interference. GPS positioning accuracy, its tracking accuracy is not high, usually in the error range of more than 1 meter; the shortcoming of tracking mode based on binocular recognition is that its computational complexity is very large, it needs a powerful processor, and its power consumption is also Higher. In addition, the current technology is not mature enough, it is easy to be disturbed by the environment and light. The tracking accuracy is not very high, it is prone to loss, and the cost is very high. It is not suitable for low-cost entertainment aerial drones. [0006] For the time being, in view of the increasingly popular aerial entertainment drones, it is urgent to find a stable, simple, low-cost, high-precision tracking method to meet the higher demands of users.

technical problem

[0007] The object of the present invention is to provide a drone tracking control system and method based on illuminating target recognition, which aims to solve the problem that the tracking accuracy of the current UAV tracking control system is not high and stable.

Problem solution

Technical solution

[0008] The present invention provides a drone tracking control system based on illuminating target recognition, including a illuminating target and a drone, and the drone includes:

[0009] at least two cameras, two cameras for performing real-time imaging of the illuminating target;

[0010] an operation processing unit, connected to the camera, identifying the illuminating target in the image captured by the camera, and calculating motion information of the illuminating target;

[0011] The flight control unit is connected to the operation processing unit, and controls the drone to follow the movement trajectory of the illuminating target according to the motion information of the illuminating target.

[0012] The present invention also provides a drone tracking control method based on illuminating target recognition, the method is based on at least one illuminating target recognition tracking control, and the method comprises the following steps:

[0013] employing at least two cameras for performing real-time imaging of the illuminating target;

[0014] identifying the illuminating target in the image captured by the camera, and calculating motion information of the illuminating target;

[0015] controlling, according to the motion information of the illuminating target, the drone to follow the moving trajectory of the illuminating target to fly.

Advantageous effects of the invention

Beneficial effect

[0016] The above-described drone tracking control system and method based on illuminating target recognition adopts two common cameras to perform real-time imaging on the illuminating target, and simultaneously calculate the distance and displacement amount of the illuminating target in the image, and exercise The direction is analyzed and judged, and the motion quantity and orientation information are output to the flight control system, and the flight control system performs corresponding motion according to the change amount, thereby implementing the tracking function. Since the position coordinates of the illuminating target can be calculated, the trajectory of the illuminating target can be determined, and the trajectory and flight control can be performed. The preset trajectory comparison in the system determines the running command of the corresponding trajectory, and achieves the purpose of controlling the drone flight by the illuminating target motion, for example, the illuminating target moves vertically upwards, indicating that the aircraft is rising, and the illuminating target is moving vertically downward. Said the aircraft is falling and so on. The invention solves the problems of instability, complicated system, high cost and low precision. In addition to more accurate tracking, it can also be used as a new control method, so that the drone only needs to add a set. The device can realize a variety of intelligent functions, which is very suitable for popular aerial entertainment drones.

Brief description of the drawing

DRAWINGS

1 is a schematic diagram of a module of a drone tracking control system based on illuminating target recognition in accordance with a preferred embodiment of the present invention;

2 is a schematic diagram of a module of an arithmetic processing unit in the UAV tracking control system based on the illuminating target recognition shown in FIG. 1;

3 is a schematic diagram of a coordinate system of a illuminating target;

4 is a schematic diagram of a module of a flight control unit in the UAV tracking control system based on the illuminating target recognition shown in FIG. 1.

5 is a flowchart of a drone tracking control method based on illuminating target recognition according to a preferred embodiment of the present invention.

6 is a flow chart of a method for calculating motion information of a illuminating target in a preferred embodiment of the present invention;

7 is a flow chart of a method for controlling a drone to track a trajectory of a illuminating target in accordance with a preferred embodiment of the present invention.

Embodiments of the invention

[0024] In order to make the technical problems, technical solutions and advantageous effects to be solved by the present invention more clearly, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

Referring to FIG. 1, a drone tracking control system based on illuminating target recognition in a preferred embodiment of the present invention includes at least one illuminating target 10 and a drone 20.

The drone 20 includes a first camera 21, a second camera 22, an arithmetic processing unit 23, and a flight control unit. twenty four.

[0027] Two cameras 21, 22 are used for performing actual imaging of the illuminating target 10; an arithmetic processing unit 23 is connected to the first camera 21 and the second camera 22, and the first camera 21 and the first camera are identified. The illumination target 10 in the image captured by the second camera 22, and calculating motion information of the illumination target 10; the flight control unit 24 is connected to the operation processing unit 23, and controls the location according to the motion information of the illumination target 10. The drone 20 follows the movement trajectory of the illuminating target.

[0028] In this embodiment, the illuminant of the illuminating target 10 at least one side (visual cross section) uniformly emits light, such as a photosphere. The reason why the photosphere is used is that the illuminating of the photosphere is very uniform, and it is difficult to find such uniformity in nature. The light source is therefore not easily disturbed by ambient light. In addition, its characteristics are very obvious, and it is easy to capture during image analysis and processing, which greatly reduces the computational complexity and difficulty of the software. In addition, since the light ball is an illuminant, it can be captured by the camera in the case of poor light at night, so that this tracking method can be used during day and night, which is more advantageous than the tracking of binocular recognition. Thus, the arithmetic processing unit 23 employs a general performance _DS P (Digital Signal Processor) having a simple image algorithm, which is sufficient to accurately calculate the direction of motion and the amount of motion of the light sphere. The flight control unit 24 is a central processing unit of the drone 20, and uses chips such as ARM, Intel, and AMD.

[0029] In a more specific embodiment, the flight control unit 24 searches for a pre-stored, corresponding predetermined flight trajectory based on the motion information of the illuminating target 10, and controls the drone 20 to fly with the predetermined flight trajectory. The predetermined flight trajectory found according to the motion information of the current illuminating target 10 is: a flight path simulating a moving trajectory of the current illuminating target, and the motion information of the flight path has a corresponding relationship with the motion information of the current illuminating target, if no The movement speed, motion acceleration, movement distance, etc. of the man-machine are n times of the illumination target, and the motion direction is the same as the illumination target, opposite or offset by the preset angle. Thus, the motion information of the illuminating target 10 includes: one or more of the moving speed, the moving distance, the moving direction, and the motion acceleration of the illuminating target 10.

[0030] In a more specific embodiment, first, the drone 20 activates the tracking or illuminating target (light ball) control mode, the flight control unit 24 notifies the arithmetic processing unit 23 to start the operation, and the arithmetic processing unit 23 activates the left camera. The module L (first camera 21) and the right camera module R (second camera 22), the two camera modules start searching for the light ball, this process requires the flight control unit 24 and the arithmetic processing unit 23 to communicate with each other, flight control Unit 24 continuously adjusts the height and direction of the drone 20 until the photosphere is captured at the optimal angle. The best way to set the distance between the photosphere and the drone 20 is 5 meters. Of course, the user can change the distance and height, orientation, etc. It is guaranteed that the dual camera can shoot the light ball at the same time.

[0031] In a specific embodiment, referring to FIG. 2, the operation processing unit 23 includes an acquisition module 231, an identification module 232, and a calculation module 233.

[0032] The obtaining module 231 is configured to continuously acquire images captured by the first camera 21 and the second camera 22, and for continuous acquisition, the inter-turn interval may be continuously acquired, or the shooting interval of the camera itself may be directly taken as an example. That is, each image needs to be acquired; it is also possible to use several images at intervals as the required elements.

[0033] The identification module 232 is configured to identify the illuminating target 10 in each of the images, and recognize the illuminating target 10 in the captured image by using an image recognition technology and mark the coordinates of the illuminating target 10 to be calculated.

[0034] The calculation module 233 is configured to calculate coordinate information of the illumination target 10 relative to the corresponding camera in each of the images, and calculate an preset time according to the coordinate information of the illumination target 10 in each of the images. The motion information of the illuminating target 10 is described. In particular, for the recognition, the arithmetic processing unit 23 needs to continuously calculate the movement trace (motion information) of the illuminating target 10, and the cumulative displacement amount of the illuminating target 10 is outputted for a period of time, and the flight control unit of the drone 20 24 responded again. For tracking control, it is limited to a preset time (same period of time), the parameters of the predetermined flight trajectory match the motion information of the preset daylighting target 10, then the tracking control is performed, if not, then Abandon tracking control, you can also fly with the default flight path, such as hovering, landing, and so on. The preset time can be 0.2, 0.5 seconds, etc.

[0035] In this embodiment, referring to FIG. 3, the coordinate information (X, Υ, Z) of the illuminating target 10 satisfies the following formula:

Z = (b * focal - length) I (x_camL - x_camR);

[0037] X = x_camL * Z I focal_length;

Y = y_camL * Z I focal_length;

[0039] wherein: the coordinate system of the illuminating target 10 is zero with the first camera 21, and in other embodiments, the other camera 22 may be selected as the zero point of the coordinate system, or other positions. The first camera 21 and the second camera 22 fall on the X axis, that is, the X axis of the coordinate system of the first camera 21 and the second camera 22; the Z axis of the coordinate system is the optical axis, and the camera The direction of pointing (the direction perpendicular to the mirror of the camera). The Y axis of the coordinate system is perpendicular to the plane in which the x and z axes are located.

[0040] More specifically, _X _camL, respectively the value y_camL emission target 10 (P point) X-axis coordinate of the first point in the imaging camera 21, Y-axis coordinate; _{X _camR} value of said target luminescent 10 is the X-axis coordinate of the imaged point of the second camera 22; since the P point is constantly changing, the values of x_camL, x_camR, and y_cam L are also constantly changing, and the values of x_camL, x_camR, and y_camL are expressed in pixel coordinates.

[0041] Method of collecting the value of x_camL: The image captured by the first camera 21, in OpenCV (the full name of OpenCV is: Open Source Computer Vision Library. OpenCV is a cross-platform computer vision library based on BSD license (幵源)) After processing, three values of XL, YL, and ZL can be obtained by the function of cvFindStereoCorresponden ceBM, and XL, YL, and ZL are respectively the imaging points of the illuminating target 10 in the first camera 21 on the X-axis, the Y-axis, and the Z-axis. The distance from a camera 21 (zero point). This day is x_camL = XL, y_camL = YL.

The method of values [0042] Acquisition of _{X _camR:} 22 second camera image acquisition, at opencv processed, a value obtained by cvFindStereoCorrespondenceBM function XR, XR is a light emitting point of the second target 10 in the imaging camera 22 on the X axis The distance from the second camera 22. Therefore, _X _camR = b + XR. The value of b is the distance between the two cameras; in this embodiment, the focal lengths of the first camera 21 and the second camera 22 are the same, and the value of focal-length is the focal length of the camera.

[0043] The operation processing unit 23 can calculate the coordinates of the photosphere in each image by the above formula and technically output the optical ball motion information of a certain interval, and transmit the information to the flight control unit 24, and the fly control system can Accurately follow the light ball, and the same purpose is to control the drone 20 through the motion trajectory by judging the trajectory of the light ball.

[0044] In a specific embodiment, referring to FIG. 4, the flight control unit 24 includes a storage module 241, a lookup module 242, and a control module 243. The storage module 241 is configured to pre-store the motion information of the illuminating target and the table data of the predetermined flight trajectory of the drone corresponding thereto; the searching module 242 is configured to check the table according to the motion information of the illuminating target 10, and obtain the corresponding information corresponding to the motion information. The drone 20 predetermines the flight trajectory; the control module 243 is configured to output the corresponding flight control command according to the found predetermined flight trajectory to control the drone 20 to fly with the preset flight information. For example: the flight control unit 24 pre-stores the following predetermined flight trajectories of the drone: the light ball moves upwards by 10 to 30 CM, correspondingly, the drone 20 rises by 0.5 m; The downward movement moves upwards by 10 to 30 CM, correspondingly, indicating that the drone 20 is lowered by 0.5 m; the light sphere is moved to the left by 10 to 30 CM, correspondingly, indicating that the drone 20 flies 1 m to the left; the photosphere is 10 to the right 30CM, correspondingly, means to fly to the right for 1 meter; the light ball is drawn in place, correspondingly, indicating that the drone 20 is shaking about 5 times, the above parameters are default values, and this value can be passed through the APP of the drone 20 The settings are the same as below. It should be noted that this example contains these kinds of control commands, but is not limited to these kinds of control commands.

[0045] The arithmetic processing unit 23 calculates the position of the light sphere in real terms, and if it is determined that there is motion, information such as the direction and speed of the motion amount is randomly transmitted to the flight control unit 24. An example of pre-storing the motion information of each illuminating target and the table data corresponding to the predetermined flight trajectory of the UAV 20 is as follows:

[0046] For example, it is detected that the light ball is again operated to the front of the unmanned person at a speed of 0.5 m/s, and the flight control unit 24 receives this information, and controls the aircraft to operate at the same speed and orientation, thus making the light ball and the drone 20 The distance between them is relatively fixed, thus achieving a more accurate tracking. Usually, if the speed of the light ball is less than 1 m / s, the following accuracy can be 10 CM.

[0047] Further, if the user holds the light ball by hand, move upward 10 to 30 CM, then move downward 10 to 30 CM, then move left to 10 to 30 CM, then right to 10 to 30 CM, and then draw the circle in place. After the calculation processing unit 23 calculates the displacement amounts, the flight control unit 24 is notified, the flight control unit 24 compares and finds the corresponding control mode, and executes the control command, and the drone 20 first moves upward by 0.5 meters and then runs downward by 0.5 meters. , Run 1 meter to the left, then 1 meter to the right, and then 5 times to the left and right. This is to achieve the purpose of controlling the drone 20 movement through the light ball, bringing a new experience to the user.

[0048] Further, in order to improve entertainment and precision, the user can also operate through two light balls, so that the defined action and control manners are more, and the two light balls are used as reference objects before each other, so that the DSP operation is performed. More accurate, which makes the recognition rate higher, can complete the identification of more difficult movements, and can bring more fun to the user.

[0049] The UAV tracking control system realizes accurate tracking by binocularly identifying a uniform specific illuminant; achieving light ball control by means of preset different motion trajectories in the flight control system to represent different control commands combined with photosphere positioning technology The drone 20 solves the problems of instability, complicated system, high cost and low precision.

[0050] Furthermore, a drone tracking control method based on illuminating target recognition is disclosed, which method is based on at least one illuminating target recognition tracking control, and the illuminating target is preferably a uniformly illuminating light sphere. [0051] In a preferred embodiment, first, the UAV starts tracking or illuminating the target (light ball) control mode, and the flight control system notifies the DSP operation processing system to start working, and the DSP starts the left camera module L and the right camera module. Group R, two camera modules start searching for the light ball. This process requires the flight control unit and the arithmetic processing unit to communicate with each other. The flight control unit continuously adjusts the height and direction of the drone until the light ball is captured at the optimal angle. (It is usually the best way to place the ball in the middle of the two cameras.) The default distance between the ball and the drone is 5 meters. Of course, you can change the distance and height, orientation, etc. The camera can shoot the light ball at the same time.

[0052] Referring to FIG. 5, in a preferred embodiment, the method includes the following steps:

[0053] Step S110, two cameras are used for performing real-time imaging of the illuminating target, and the two cameras include a first camera and a second camera. The focal lengths of the two cameras are the same.

[0054] Step S120: Identify the illuminating target in the image captured by the camera, and calculate motion information of the illuminating target. The motion information of the illuminating target includes: one or more of a moving speed, a moving path, a moving direction, a moving time, and a moving acceleration of the illuminating target.

[0055] Step S130, controlling the drone to follow the movement trajectory of the illuminating target according to the motion information of the illuminating target.

[0056] The tracking and simple control system formed by the two cameras and the DSP computing system plus a specific illuminant can greatly enhance the entertainment and intelligence of the drone. The core elements are two points. The first is The light ball, the reason why the light ball is used is that the light ball is very uniform, and it is difficult to find such a uniform light source in nature, so it is not easily interfered by ambient light. In addition, its characteristics are very obvious, and the image analysis processing is performed. The process is easy to capture, greatly reducing the amount of computation and difficulty of the software. Using the general performance DSP and simple image algorithm, the direction and amount of movement of the light sphere can be accurately calculated.

[0057] In a more specific embodiment, the flight control system controls the drone to follow the movement trajectory of the illuminating target according to the motion information of the illuminating target to: search for pre-stored according to the motion information of the illuminating target, Corresponding predetermined flight trajectory, and controlling the drone to fly with the predetermined flight trajectory. The predetermined flight trajectory found according to the motion information of the current illuminating target is: a flight path simulating a moving trajectory of the current illuminating target, and the motion information of the flight path has a corresponding relationship with the motion information of the current illuminating target, such as no one. The motion speed, motion acceleration, motion distance, etc. of the machine are n times of the illumination target, and the motion direction is the same as the illumination target, or opposite to the preset angle. [0058] In a more detailed embodiment, referring to FIG. 6 and FIG. 3, step S120 specifically includes:

[0059] Step S121, the images captured by the first camera 21 and the second camera 22 are continuously acquired. For continuous acquisition, you can preset the inter-turn interval to obtain continuously, or you can directly take the camera's own shooting interval as an example, that is, each image needs to be acquired; or several images per interval can be used as the required elements.

[0060] Step S122, identifying the illuminating target in each of the images. The image recognition technology is used to identify the illuminating target in the captured image and mark it to the coordinates of the technical illuminating target.

[0061] Step S123: Calculate coordinate information of the illuminating target relative to the corresponding camera in each of the images.

In this embodiment, the coordinate information (x, Υ, ζ) of the illuminating target satisfies the following formula:

006 = (b *focal_length) I (x_camL - x_camR);

X = x_camL * Z I focal_length;

Y = y_camL * Z I focal_length;

[0065] wherein: the coordinate system of the illuminating target is zero point with the first camera 21, and in other embodiments, the other camera may be selected as the zero point of the coordinate system, or other positions. The first camera 21 and the second camera 22 are on the X-axis, that is, the X-axis of the coordinate system of the first camera 21 and the second camera 22; the Z-axis of the coordinate system is the optical axis, and the camera is pointed The direction (the direction perpendicular to the mirror of the camera). The Y axis of the coordinate system is perpendicular to the plane of the X and Z axes.

[0066] More specifically, _X _camL, y_camL values are X axis coordinate of the light emitting point in the imaging target (P point) of the first camera 21, Y-axis coordinate; x_camR value is the target of light emission The X-axis coordinate of the imaging point of the two cameras 22; Since the P point is constantly changing, the values of x_camL, x_camR, and y_camL are also constantly changing, and the values of x_camL, x_camR, and y_camL are expressed in pixel coordinates.

[0067] Method of collecting the value of x_camL: The image captured by the first camera 21, in OpenCV (the full name of OpenCV is: Open Source Computer Vision Library. OpenCV is a cross-platform computer vision library based on BSD license (幵源) release) After processing, three values of XL, YL, and ZL can be obtained by the function of cvFindStereoCorresponden ceBM, and XL, YL, and ZL are the imaging points of the illuminating target in the first camera 21 on the X-axis, the Y-axis, and the Z-axis, respectively. The distance from the camera 21 (zero point). This day is x_camL = XL, y_camL = YL.

The method of values [0068] Acquisition of _{X _camR:} 22 second camera image acquisition, at opencv processed, a value obtained by cvFindStereoCorrespondenceBM function XR, XR is a light emitting target at a second camera The imaged point of the head 22 is on the X-axis from the second camera 22. Therefore, _X _camR = b + XR. The value of b is the distance between the two cameras; in this embodiment, the focal lengths of the first camera 21 and the second camera 22 are the same, and the value of focal-length is the focal length of the camera.

[0069] Step S124: Calculate motion information of the illumination target in the preset time according to the coordinate information of the illumination target in each of the images. Through the above formula, the coordinates of the photosphere in each image can be calculated and the optical ball motion information of a certain interval can be outputted by the technology, and the information can be transmitted to the flight control system, and the flight control system can accurately follow the light ball. By judging the trajectory of the light sphere, the purpose of controlling the drone through the motion trajectory is realized. In particular, for the recognition, it is necessary to continuously calculate the movement trace (motion information) of the illuminating target, and the cumulative displacement of the illuminating target is outputted for a period of time, and the flight control system of the drone responds again. For tracking control, it is limited to a preset time (same interval), the parameters of the predetermined flight trajectory match the motion information of the preset daytime illuminating target, then the tracking control is performed, and if it does not match, the data is discarded. Tracking control, you can also fly with the default flight path, such as hovering, landing, and so on. The preset time can be 0.2, 0.5 seconds, etc.

[0070] In a more detailed embodiment, referring to FIG. 7, searching for a pre-stored, corresponding predetermined flight trajectory according to the motion information of the illuminating target, and controlling the drone to fly the predetermined flight trajectory specifically includes:

[0071] Step S131, pre-storing the motion information of the illuminating target and the unmanned flight scheduled trajectory data table corresponding thereto.

[0072] Step S132: Obtain a predetermined flight trajectory of the UAV corresponding to the motion information according to the motion information of the illuminating target.

[0073] Step S133: Output the corresponding flight control command according to the found predetermined flight trajectory to control the non-human machine to fly with the preset flight information.

[0074] An example of pre-storing table data corresponding to the motion information of the respective light targets and the predetermined flight trajectory of the drone is as follows:

[0075] For example, it is detected that the light ball is operated to the front of the unmanned person at a speed of 0.5 m/s. The flight control system receives this information and controls the aircraft to operate at the same speed and orientation, thus causing the ball between the light ball and the drone. The distance is relatively fixed, thus achieving a more accurate tracking. Usually, if the speed of the light ball is less than 1 m / sec, the following accuracy can be 10 CM.

[0076] Further, the following motion trajectories are defined in the flight control system: The light ball moves upward 10 to 30 CM Indicates that the drone rises by 0.5 m; the downward movement of the light ball moves upwards by 10 to 30 CM, indicating that the drone is lowered by 0.5 m; the movement of the photosphere to the left by 10 to 30 CM means 1 m to the left; and the photosphere to the right by 10 to 30 CM. Flying right 1 meter; The light ball in-situ circle indicates that the drone is shaking 5 times. The above parameters are default values. This value can be set by the drone's APP. The same is true. It should be noted that this example contains these kinds of control commands, but is not limited to these kinds of control commands.

[0077] Further, if the user holds the light ball by hand, move upward 10 to 30 CM, then move 10 to 30 CM downward, then move 10 to 30 CM to the left, then 10 to 30 CM to the right, and then circle the original. After the DSP calculates these displacements, it will notify the flight control system. The flight control system compares and finds the corresponding control mode. When the control command is executed, the drone will move upwards by 0.5 meters, then run down 0.5 meters, and then run to the left. 1 meter, then run 1 meter to the right, then shake it 5 times. This is to achieve the purpose of controlling the drone movement through the light ball, bringing a new experience to the user.

[0078] Further, in order to improve entertainment and precision, the user can also operate through two light balls, so that the defined actions and control methods are more, and the two light balls are used as reference objects before each other, so that the DSP operation is performed. More accurate, which makes the recognition rate higher, can complete the identification of more difficult movements, and can bring more fun to the user.

[0079] The DSP system will calculate the position of the light sphere, and if it is determined that there is motion, information such as the direction and speed of the motion amount is randomly transmitted to the flight control system. Accurate tracking is achieved by binocular recognition of a specific illuminant; precise control is achieved by presetting different motion trajectories in the flight control system to represent different control commands in combination with the photosphere positioning technology to realize the optical ball control drone. Unstable, complex system, high cost, and low accuracy.

[0080] It will be clearly understood by those skilled in the art that, for convenience and brevity of description, only the division of each functional unit described above is exemplified. In practical applications, the above functions may be assigned to different functional units as needed. Upon completion, the internal structure of the device is divided into different functional units or modules to perform all or part of the functions described above. Each functional unit in the embodiment may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit, and the integrated unit may be implemented in the form of hardware. , can also be implemented in the form of software functional units. In addition, the specific names of the functional units are only for convenience of distinguishing from each other, and are not intended to limit the scope of protection of the present application. The specific working process of the unit in the above device, Reference may be made to the corresponding process in the foregoing method embodiments, and details are not described herein again.

[0081] In summary, the control system and method of the embodiment of the present invention adopts two common cameras to perform real-time imaging on the illuminating target, and simultaneously calculate the distance and displacement amount of the illuminating target in the image, and the moving direction. The analysis and judgment are made, and the motion quantity and the position information are output to the flight control system, and the flight control system performs corresponding motion according to the change amount, thereby implementing the tracking function. Since the position coordinates of the illuminating target can be calculated, the trajectory of the illuminating target can be determined, and the trajectory is compared with the preset trajectory in the flight control system, thereby finding the running instruction of the corresponding trajectory, and achieving the movement through the illuminating target. The purpose of controlling the flight of the drone, such as the vertical movement of the illuminating target, indicates that the aircraft is rising, the illuminating target is moving vertically downward, indicating that the aircraft is descending, and the like. The invention solves the problems of instability, complicated system, high cost and low precision. In addition to more accurate tracking, it can also be used as a new control method, so that the drone only needs to add a set. The device can realize a variety of intelligent functions, which is very suitable for popular aerial entertainment drones.

[0082] Those of ordinary skill in the art will appreciate that the elements and algorithm steps of the various examples described in connection with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the solution. A person skilled in the art can use different methods for implementing the described functions for each particular application, but such implementation should not be considered to be beyond the scope of the present invention.

[0083] In the embodiments provided by the present invention, it should be understood that the disclosed apparatus and method may be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the modules or units is only a logical function division, and the actual implementation may have another division manner, for example, multiple units or components may be Combined or can be integrated into another system, or some features can be ignored, or not executed. Alternatively, the coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interface, device or unit, and may be in electrical, mechanical or other form.

[0084] The unit described as a separate component may or may not be physically distributed, and the component displayed as a unit may or may not be a physical unit, that is, may be located in one place, or may be distributed to multiple On the network unit. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of the embodiment. In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above integrated unit can be implemented in the form of hardware or in the form of a software functional unit.

[0086] The integrated unit, if implemented in the form of a software functional unit and sold or used as a standalone product, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the embodiments of the present invention may contribute to the prior art or all or part of the technical solution may be embodied in the form of a software product stored in a storage. The medium includes a plurality of instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) or a processor to perform all or part of the steps of the methods of the various embodiments of the embodiments of the present invention. The foregoing storage medium includes: a U disk, a mobile hard disk, a read-only memory (R 0M, Read-Only Memory), a random access memory (RAM), a magnetic disk or an optical disk, and the like, which can store program codes. medium.

The above is only the preferred embodiment of the present invention, and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection of the present invention. Within the scope.

Claims

Claim

[Attachment 1] A UAV tracking control system based on illuminating target recognition, comprising: a human-machineless machine and at least one illuminating target, the drone comprising:

a first camera and a second camera, configured to perform actual imaging on the illuminating target; an arithmetic processing unit connected to the first camera and the second camera to identify images captured by the first camera and the second camera The illuminating target in the middle, and calculating motion information of the illuminating target;

The flight control unit is connected to the operation processing unit, and controls the drone to follow the movement trajectory of the illuminating target according to the motion information of the illuminating target.

[Claim 2] The illuminating target recognition-based drone tracking control system according to claim 1, wherein the illuminating target is an illuminant that emits light uniformly on at least one side.

[Claim 3] The illuminating target recognition-based drone tracking control system according to claim 1, wherein the arithmetic processing unit comprises:

An acquiring module, configured to continuously acquire images captured by the first camera and the second camera; and an identifying module, configured to identify the illuminating target in each of the images;

And a calculation module, configured to calculate coordinate information of the illumination target relative to the corresponding camera in each of the images, and calculate, according to the coordinate information, a motion f π of the illumination target in the preset time.

[Claim 4] The illuminating target recognition-based drone tracking control system according to claim 3, wherein the coordinate information (Χ, Υ, Ζ) of the illuminating target satisfies the following formula: Ζ = (b *focal_length) I (x_camL - x_camR);

X = x_camL * Z / focal_length;

Y = y_camL * Z / focal_length;

Wherein: the coordinate system of the illuminating target is the zero point of the coordinate system of one of the cameras, the straight line of the two cameras is the X axis of the coordinate system, and the z axis of the coordinate system is the direction pointed by the two cameras, the coordinate system the Y-axis and X-axis and the Z-axis perpendicular to the plane; _X _camL, the value of y _camL are X-axis coordinate of the emission point of the first target in the imaging of the camera, Y-axis coordinate; x_camR value of said target luminescent The X-axis of the second camera imaging point The value of focaUength is the focal length of the two cameras.

[Claim 5] The illuminating target recognition-based drone tracking control system according to claim 1, wherein the flight control unit controls the drone to follow the motion information according to the motion information of the illuminating target The moving trajectory of the illuminating target is specifically: searching for a pre-stored, corresponding predetermined flight trajectory according to the motion information of the illuminating target, and controlling the drone to fly with the predetermined flight trajectory.

[Claim 6] The illuminating target recognition-based drone tracking control system according to claim 5, wherein the flight control unit comprises:

a storage module, configured to pre-store motion information of the illuminating target and a data table of a predetermined flight trajectory of the unmanned aircraft;

a search module, configured to obtain a predetermined flight trajectory of the drone corresponding to the motion information according to the motion information of the illuminating target;

And a control module, configured to output, according to the found predetermined flight trajectory, a corresponding flight control command to control the drone to fly with the preset flight information.

The illuminating target recognition-based drone tracking control system according to any one of claims 1 to 6, wherein the illuminating target motion information comprises: a moving speed of the illuminating target, One or more of a moving path, a moving direction, a moving day, and an acceleration of motion.

[Attachment 8] A drone tracking control method based on illuminating target recognition, wherein the method is based on at least one illuminating target recognition tracking control, the method comprising the following steps: using two cameras for aligning Performing a real imaging of the illuminating target, the two cameras including a first camera and a second camera;

Identifying the illuminating target in the image captured by the camera, and calculating motion information of the illuminating target;

The drone is controlled to follow the moving trajectory of the illuminating target according to the motion information of the illuminating target.

The method according to claim 8, wherein the illuminating target is an illuminant that emits light uniformly on at least one surface. The illuminating target recognition-based drone tracking control method according to claim 8, wherein the illuminating target in the image captured by the camera is recognized, and motion information of the illuminating target is calculated, specifically include:

And continuously acquiring images captured by the first camera and the second camera;

Identifying the illuminating target in each of the images;

Calculating coordinate information of the illuminating target relative to the corresponding camera in each of the images; calculating motion information of the illuminating target within the preset time according to coordinate information of the illuminating target in each of the images.

The illuminating target recognition-based drone tracking control method according to claim 10, wherein the coordinate information (Χ, Υ, Ζ) of the illuminating target satisfies the following formula: Ζ = (b * focal_length) I (x_camL - x_camR);

X = x_camL * Z / focal_length;

Y = y_camL * Z / focal_length;

Wherein: the coordinate system of the illuminating target is the zero point of the coordinate system of one of the cameras, the straight line of the two cameras is the X axis of the coordinate system, and the z axis of the coordinate system is the direction pointed by the two cameras, the coordinate system The Y axis is perpendicular to the plane in which the X axis and the Z axis are located; the values of _X _camL and y _camL are the X-axis coordinate and the Y-axis coordinate of the imaging point of the illuminating target at the first camera, respectively; the value of x_camR is the illuminating target at The X-axis coordinate of the second camera imaging point; the value of focaUength is the focal length of the two cameras.

The illuminating target recognition-based drone tracking control method according to claim 8, wherein the controlling the movement trajectory of the drone following the illuminating target according to the motion information of the illuminating target is specifically :

Searching for a pre-stored, corresponding predetermined flight trajectory according to the motion information of the illuminating target, and controlling the drone to fly with the predetermined flight trajectory.

The illuminating target recognition-based drone tracking control method according to claim 12, wherein the searching for a pre-stored, corresponding predetermined flight trajectory according to the motion information of the illuminating target, and controlling the unmanned person The step of flying the machine on the predetermined flight trajectory is specifically Pre-storing motion information of the illuminating target and a data table of a predetermined flight trajectory of the drone corresponding thereto;

Obtaining a predetermined flight trajectory of the drone corresponding to the motion information according to the motion information of the illuminating target;

And outputting the corresponding flight control command according to the found predetermined flight trajectory to control the drone to fly with the preset flight information.

The method for controlling the tracking of the vehicle based on the illuminating target according to any one of claims 8 to 13, wherein the motion information of the illuminating target comprises: a moving speed of the illuminating target, One or more of a moving path, a moving direction, a moving day, and an acceleration of motion.