Tracking holder for visual dynamic capture of unmanned aerial vehicle
By designing visual dynamic capture and tracking gimbal and simulation environment on drones, the camera stability problem of drones during fast movement or external interference is solved, high-quality image and video capture is achieved, and the drone's Autonomous flight and intelligent control capabilities.
Patent Information
- Application Number
- CN202411398304.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-10-09
- Publication Date
- 2025-01-03
AI Technical Summary
Existing drone technology is difficult to maintain stable tracking of the camera to the target when it moves quickly or is subject to external interference, affecting image and video quality.
A tracking gimbal for drone visual dynamic capture is designed to ensure that the camera always points to the target through precise mechanical design and control systems. It combines technologies such as Cygwin, MAVProxy, SITL, dronekit-python and MissionPlanner to provide simulation and simulation environments. Carry out tests and controls to achieve autonomous flight and target tracking of drones.
Improve the quality of images and videos, ensure stable flight and target tracking of drones in complex environments, reduce development and testing risks, and enhance the flexibility and intelligence of drones.
Smart Images

Figure CN119240026A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of unmanned aerial vehicles, in particular to a tracking gimbal for visual dynamic capture of unmanned aerial vehicles. Background Art
[0002] Drones are aircraft that are controlled by electronic remote control devices and flight control systems, and do not require manual control by the pilot. There are many types of drones, such as multi-rotor drones and fixed-wing drones. The mainstream in the market is the lightweight, flexible and relatively low-priced quad-rotor drone. Currently, consumer drones are mainly dominated by DJI's DJI Mavic, Mini and other series. The continuous development of drone hardware, the continuous optimization of flight control system design and the rapid development of image intelligence algorithms have made people devote themselves to deploying algorithms on drones in recent years, allowing drones to complete some tasks or achieve goals that are difficult for people to complete.
[0003] The onboard camera gives the drone a pair of eyes, and the target tracking algorithm gives the drone a brain, allowing the drone to autonomously complete some tasks that are difficult to complete with human labor. For example, at the scene of an earthquake disaster, the drone can be equipped with search and rescue equipment to complete the search mission in the dangerous area, greatly reducing the difficulty of rescue workers, speeding up the dispatch of search and rescue resources, and also ensuring the safety of personnel; in terms of power inspection, for power workers, daily inspection work in steep mountainous areas is difficult, and the use of drones can ignore the influence of geographical elements, autonomously complete the inspection work, and return safely, ensuring the personal safety of power workers; in exciting sports events, the athletes' wonderful performance moments may be difficult to capture with ordinary cameras, but the use of drones can always maintain the best angle for tracking flight and record every wonderful moment of the athletes. To this end, we have a drone visual dynamic capture tracking gimbal to solve the above problems. Summary of the invention
[0004] The purpose of the present invention is to provide a tracking gimbal for visual dynamic capture of unmanned aerial vehicles to solve the problems raised in the above-mentioned background technology.
[0005] The technical solution of the present invention is: a tracking gimbal for visual dynamic capture of a drone, including a drone power system, a drone sensor design and a drone tracking gimbal design. The power system of the drone includes a drone frame, a motor is fixedly installed on the drone frame, an electric regulator is fixedly installed on the motor, a propeller is fixedly connected to the drone frame, the motor is electrically connected to a battery, and a sensor is arranged on the drone frame.
[0006] Preferably, a gyroscope is fixedly mounted on the sensor, a magnetic compass is fixedly mounted on the sensor, and a barometer is fixedly mounted on the UAV frame.
[0007] Preferably, the output end of the motor is fixedly connected to the propeller, and the motor is electrically connected to the sensor.
[0008] Preferably, the power system design of the UAV includes the following steps: When installing the motor of the drone, you need to ensure that one end of the motor is fixed to the motor seat of the frame arm and the other end is fixed to the propeller. The motor needs to work with the ESC. 1) ESC: ESC is mainly used to control the speed of the motor. The power system needs to be equipped with a suitable ESC. 2) Propeller: The propeller is the component that directly generates thrust. When the motor drives the propeller to rotate, it will generate a reverse torque, causing the frame to rotate in the opposite direction. By rotating one motor forward and one motor backward, the reverse torque can be offset, and finally lift is generated to help the drone take off. 3) Battery: The drone battery mainly uses a 6000mA lithium battery.
[0009] Preferably, the design of the drone sensor includes the following steps: Step S1: The main function of the accelerometer is to provide the acceleration force of the drone. The direction of the force is the XYZ three axes, which determines the tilt angle of the drone when it is stationary. The accelerometer is also used to provide horizontal and vertical linear acceleration, which can be used to monitor the vibration of the drone. Step S2: The gyroscope monitors the angular velocity of the XYZ axes and calculates the change rate of the pitch, roll and yaw angles. The change of the angle information obtained by the gyroscope can be used to maintain the stability of the drone and prevent shaking. The gyroscope can also ensure that the drone rotates according to the angle set by the user control device; Step S3: The function of the magnetic compass is to provide a sense of direction for the drone. It can provide data on the magnetic field to which the device is subjected in each axis of XYZ. Then the algorithm of the microcontroller will process the relevant magnetic field data to provide the heading angle related to the magnetic north pole. The final detection of the geographic orientation is achieved by relying on these data. Step S4: In the application of drones, the magnetic compass is not only used to sense the direction, but also can be used to detect the surrounding magnetism and ferrous metals, such as electrodes, wires, vehicles, and other drones, to avoid flight accidents; Step S5: The barometer is used to calculate the current altitude information through the atmospheric pressure. The pressure sensor can detect the atmospheric pressure of the earth. The data provided by the barometer can assist the drone in navigation and flying to a predetermined altitude. Step S6: GPS provides the user with the specific location information of the drone, and the location information of the drone can be monitored through signals even if it is out of the field of view.
[0010] Preferably, the steps of designing the drone tracking gimbal are as follows: Step S1: Install Cygwin, which is a UNIX-like simulation environment running on the Windows platform, and add CYGWIN_HOME to the system environment variables; Step S2: Install MAVProxy, install the executable file and accept the license and all default installation options; Step S3: Run SITL, open a Cygwin64 terminal, run SITL to start the simulation program, then SITL and MAVProxy will be opened, and MAVProxy will open three windows: the first one can input commands directly to SITL, the second one can display real-time flight control status and information, and the third one is a map showing the current location of the drone, and you can control the movement of the drone and perform flight missions through commands / right-clicking the mouse; Step S4: Send instructions to SITL to control the drone. Enter the command modeguided;armthrottle;takeoff40; here, the drone will switch to the guided mode, increase the throttle, and take off to a height of 40 meters. Then enter the following command rc31500;modecircle;paramsetcircle_radius2000;". At this time, the drone will start the circle mode and set the flight radius to 2000 cm, that is, 20 meters. If you want the drone to return successfully, you only need to enter the command modertl: 1) Download dronekit-python, compile and install dronckit, and install python2.7; 2) Run SITL and MAVProxy; 3) Install MissionPlanner, which is the drone ground station and can also be used to simulate the flight control connection, here through UDP connection; Step S5: Run the flight control script. First, use UDP to connect to the SITL simulator through the local port 14450: Then perform a pre-takeoff check on the drone and unlock the drone: Next, switch the drone flight mode to "GUIDED" mode, confirm the motor unlock status, send a takeoff command, the drone takes off and flies to the target altitude, and continue to send commands to let the drone fly to the first set waypoint. The waypoint can be set to a certain longitude and latitude through the parameter LocationGlobalRelative, and then send a return command. The drone will automatically return to the top of the home point and then automatically land to complete the flight mission. Finally, the output of the flight control module is the control command. The next step is to transmit the control command to the flight control, and then control the drone to autonomously track the target according to the calculation results in OFFBOARD mode.
[0011] The present invention provides a tracking gimbal for visual dynamic capture of unmanned aerial vehicles through improvement, which has the following improvements and advantages compared with the prior art: First, the stability and accuracy of the gimbal are crucial to obtaining high-quality images or videos. Through precise mechanical design and control system, the gimbal can ensure that the camera always points to the target and remains stable even when the drone moves quickly or is subject to external interference.
[0012] Second: In this invention, by using Cygwin, developers can simulate the UNIX environment on the Windows platform, so that they can develop and test the UAV flight control system on a wider range of operating systems. SITL (Software-In-The-Loop) simulation allows developers to test the UAV code and flight control logic before the actual flight, reducing the risk of hardware damage and flight accidents caused by code errors. The three window interfaces provided by MAVProxy enable developers to monitor the status of the UAV in real time, send control commands, and intuitively control the movement and flight missions of the UAV through the map window, improving development efficiency and user experience. Using dronekit-python and python2.7, developers can write scripted flight control programs and communicate with SIT through the programming interface. L simulator interaction to achieve complex control logic for drones. By installing drone ground station software such as MissionPlanner, developers can simulate real flight control and mission planning processes in a simulation environment to further verify the correctness of flight logic and mission execution. Using scripts and ground station software, developers can define highly customized flight missions, including setting flight modes, flight altitudes, flight paths, etc., providing greater flexibility for drone applications. Each component in the entire process is customizable and can be easily integrated into a larger drone system, or used in conjunction with other development tools and technology stacks. The final output of the flight control module is control instructions, which can be further transmitted to real flight control hardware, laying the foundation for autonomous flight and target tracking in OFFBOARD mode. BRIEF DESCRIPTION OF THE DRAWINGS
[0013] The present invention will be further explained below in conjunction with the accompanying drawings and embodiments: Figure 1 is a schematic diagram of the UAV power system of the present invention; Figure 2 It is a flight control module algorithm flow chart of the present invention. DETAILED DESCRIPTION
[0014] The present invention is described in detail below, and the technical solutions in the embodiments of the present invention are clearly and completely described. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0015] The present invention provides a tracking gimbal for visual dynamic capture of a drone through improvement. The technical solution of the present invention is: like Figure 1 - Figure 2As shown, a tracking gimbal for visual dynamic capture of a UAV includes a UAV power system, a UAV sensor design and a UAV tracking gimbal design. The power system of the UAV includes a UAV frame, a motor is fixedly installed on the UAV frame, an electric regulator is fixedly installed on the motor, a propeller is fixedly connected to the UAV frame, the motor is electrically connected to a battery, and a sensor is arranged on the UAV frame.
[0016] Furthermore, a gyroscope is fixedly mounted on the sensor, a magnetic compass is fixedly mounted on the sensor, and a barometer is fixedly mounted on the UAV frame, which together provide UAV developers with an efficient, safe, and intelligent development and testing environment, enabling UAVs to cope with various complex environments and flight missions, and promoting the continuous development of UAV technology.
[0017] Furthermore, the output end of the motor is fixedly connected to the propeller, and the motor is electrically connected to the sensor. This not only improves the transmission efficiency and flight stability, but also enhances safety, realizes rapid replacement and maintenance, and has the ability of intelligent monitoring and adjustment, adapting to the needs of various application scenarios.
[0018] Furthermore, the power system design of the UAV includes the following steps: When installing the motor of the drone, you need to ensure that one end of the motor is fixed to the motor seat of the frame arm and the other end is fixed to the propeller. The motor needs to work with the ESC. 1) ESC: ESC is mainly used to control the speed of the motor. The power system needs to be equipped with a suitable ESC. 2) Propeller: The propeller is the component that directly generates thrust. When the motor drives the propeller to rotate, it will generate a reverse torque, causing the frame to rotate in the opposite direction. By rotating one motor forward and one motor backward, the reverse torque can be offset, and finally lift is generated to help the drone take off. 3) Battery: The drone battery mainly uses a 6000mA lithium battery. The gyroscope is mainly used to measure and maintain the angular velocity and angular displacement of the drone. It can help the drone maintain a stable attitude during flight, especially when it is subject to external interference (such as wind, mechanical vibration, etc.). The data of the gyroscope is usually used in conjunction with the drone's flight control system to achieve automatic stabilization, attitude control and navigation. The magnetic compass is used to measure the earth's magnetic field to determine the direction of the drone (i.e., heading). It is not affected by external acceleration, so it can provide stable heading information. The data of the magnetic compass is crucial for the navigation of the drone, especially when there is no GPS signal or the GPS signal is unstable. It can help the drone determine its direction relative to the earth's magnetic north pole for accurate navigation. The barometer estimates the altitude of the drone by measuring atmospheric pressure. Since atmospheric pressure changes with altitude, the barometer can provide a relatively accurate altitude measurement. The barometer plays a key role in the flight control of the drone, especially when performing operations such as vertical take-off and landing, hovering and altitude holding. It can help the drone accurately sense its height relative to the ground, thereby performing precise altitude control. The benefit of installing sensors such as gyroscopes, magnetic compasses and barometers on drones is that they can provide key information about the drone's attitude, heading and altitude, thereby helping the drone achieve stable and accurate flight and navigation.
[0019] Furthermore, the design of the drone sensor includes the following steps: Step S1: The main function of the accelerometer is to provide the acceleration force of the drone. The direction of the force is the XYZ three axes, which determines the tilt angle of the drone when it is stationary. The accelerometer is also used to provide horizontal and vertical linear acceleration, which can be used to monitor the vibration of the drone. Step S2: The gyroscope monitors the angular velocity of the XYZ axes and calculates the change rate of the pitch, roll and yaw angles. The change of the angle information obtained by the gyroscope can be used to maintain the stability of the drone and prevent shaking. The gyroscope can also ensure that the drone rotates according to the angle set by the user control device; Step S3: The function of the magnetic compass is to provide a sense of direction for the drone. It can provide data on the magnetic field to which the device is subjected in each axis of XYZ. Then the algorithm of the microcontroller will process the relevant magnetic field data to provide the heading angle related to the magnetic north pole. The final detection of the geographic orientation is achieved by relying on these data. Step S4: In the application of drones, the magnetic compass is not only used to sense the direction, but also can be used to detect the surrounding magnetism and ferrous metals, such as electrodes, wires, vehicles, and other drones, to avoid flight accidents; Step S5: The barometer is used to calculate the current altitude information through the atmospheric pressure. The pressure sensor can detect the atmospheric pressure of the earth. The data provided by the barometer can assist the drone in navigation and flying to a predetermined altitude. Step S6: GPS provides the user with the specific location information of the drone. Even if it is out of the field of view, the location information of the drone can be monitored through signals. The combination of the accelerometer and the gyroscope can accurately monitor the attitude changes of the drone, such as tilt, pitch, roll and yaw, so as to ensure that the drone remains stable during flight and avoids loss of control due to external forces. The magnetic compass provides the drone with a sense of direction, allowing the drone to accurately fly along the predetermined route and prevent collisions with other objects (such as wires, vehicles, other drones), thereby improving flight safety. The barometer can calculate the altitude information of the current position and assist the drone in accurate vertical navigation, so that it can accurately rise to a predetermined altitude or maintain a stable flight altitude. GPS provides The global positioning information of the drone can be obtained, and the position of the drone can be tracked even out of sight, which makes remote control and automatic flight of the drone possible. Through the data fusion of these sensors, the drone can realize autonomous flight, such as automatic return, autonomous hovering and other advanced functions. Combined with image processing and other technologies, the drone can also realize intelligent obstacle avoidance, that is, according to the obstacle information detected by the sensor, the flight route is automatically adjusted to avoid collision. Since the drone can fly stably and accurately, and has navigation and positioning capabilities, it can be applied to a variety of scenarios, such as aerial photography, terrain surveying and mapping, agricultural plant protection, logistics and transportation, etc. Through precise control and stable flight, the drone can provide a smoother, high-quality shooting and video experience to meet users' needs for high-quality images.
[0020] Furthermore, the steps of designing the drone tracking gimbal are as follows: Step S1: Install Cygwin, which is a UNIX-like simulation environment running on the Windows platform, and add CYGWIN_HOME to the system environment variables; Step S2: Install MAVProxy, install the executable file and accept the license and all default installation options; Step S3: Run SITL, open a Cygwin64 terminal, run SITL to start the simulation program, then SITL and MAVProxy will be opened, and MAVProxy will open three windows: the first one can input commands directly to SITL, the second one can display real-time flight control status and information, and the third one is a map showing the current location of the drone, and you can control the movement of the drone and perform flight missions through commands / right-clicking the mouse; Step S4: Send instructions to SITL to control the drone. Enter the command modeguided;armthrottle;takeoff40; here, the drone will switch to the guided mode, increase the throttle, and take off to a height of 40 meters. Then enter the following command rc31500;modecircle;paramsetcircle_radius2000;". At this time, the drone will start the circle mode and set the flight radius to 2000 cm, that is, 20 meters. If you want the drone to return successfully, you only need to enter the command modertl: 1) Download dronekit-python, compile and install dronckit, and install python2.7; 2) Run SITL and MAVProxy; 3) Install MissionPlanner, which is the drone ground station and can also be used to simulate the flight control connection, here through UDP connection; Step S5: Run the flight control script. First, use UDP to connect to the SITL simulator through the local port 14450: Then perform a pre-takeoff check on the drone and unlock the drone: Next, switch the drone flight mode to "GUIDED" mode, confirm the motor unlock status, send a takeoff command, the drone takes off and flies to the target altitude, and continue to send commands to let the drone fly to the first set waypoint. The waypoint can be set to a certain longitude and latitude through the parameter LocationGlobalRelative, and then send a return command. The drone will automatically return to the top of the home point and then automatically land to complete the flight mission. The final output of the flight control module is the control command. The next step is to transmit the control command to the flight control, and then control the drone to track the target autonomously according to the calculation results in OFFBOARD mode. Provide a UNIX-like environment (through Cygwin) that allows developers to easily use tools and software designed for Linux or UNIX on Windows. MAVProxy and SITL provide drone flight simulation functions, allowing developers to test flight control scripts and algorithms before actual flight, thereby reducing the risk in actual flight. Developers can simulate various flight conditions and scenarios, such as different wind speeds, altitudes, etc. The flight control script allows developers to precisely control the behavior of the drone, including takeoff, flight, mission execution, return, and landing. The drone can fly autonomously according to preset waypoints without human intervention, thereby improving flight efficiency and accuracy. When using the drone for complex tasks such as target tracking, autonomous navigation and decision-making can be achieved through algorithms and sensor data, which improves the intelligence level of the drone. The gyroscope provides information about the angular velocity and angular displacement of the drone, which helps to achieve stable flight and attitude control of the drone. The magnetic compass provides stable heading information, which is essential for the navigation of the drone, especially when the GPS signal is unstable or there is no GPS signal. The barometer provides information about the altitude of the drone, which helps to achieve precise altitude control and vertical take-off and landing operations. The fusion of these sensor data can greatly improve the flight performance and stability of the drone, enabling it to cope with various complex environments and flight missions. By using a simulated environment and flight control scripts, developers can develop and test drones more efficiently, shorten the development cycle and reduce costs. The sensors installed on the drone provide key information about the attitude, heading, and altitude of the drone, enabling the drone to achieve stable and accurate flight and navigation.
[0021] The above description enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.
Claims
1. A tracking gimbal for visual dynamic capture of unmanned aerial vehicles, including a power system of unmanned aerial vehicles, a sensor design of unmanned aerial vehicles and a tracking gimbal design of unmanned aerial vehicles, characterized in that: The power system of the drone comprises a drone frame, a motor is fixedly mounted on the drone frame, an electric regulator is fixedly mounted on the motor, a propeller is fixedly connected to the drone frame, the motor is electrically connected to a battery, and a sensor is arranged on the drone frame.
2. The tracking gimbal for visual dynamic capture of an unmanned aerial vehicle according to claim 1, characterized in that: A gyroscope is fixedly mounted on the sensor, a magnetic compass is fixedly mounted on the sensor, and a barometer is fixedly mounted on the UAV frame.
3. The tracking gimbal for visual dynamic capture of an unmanned aerial vehicle according to claim 1, characterized in that: The output end of the motor is fixedly connected to the propeller, and the motor is electrically connected to the sensor.
4. The tracking gimbal for visual dynamic capture of an unmanned aerial vehicle according to claim 1, characterized in that: The power system design of the UAV includes the following steps: When installing the motor of the drone, you need to ensure that one end of the motor is fixed to the motor seat of the frame arm and the other end is fixed to the propeller. The motor needs to work with the ESC. 1) ESC: ESC is mainly used to control the speed of the motor. The power system needs to be equipped with a suitable ESC. 2) Propeller: The propeller is the component that directly generates thrust. When the motor drives the propeller to rotate, it will generate a reverse torque, causing the frame to rotate in the opposite direction. By rotating one motor forward and one motor backward, the reverse torque can be offset, and finally lift is generated to help the drone take off. 3) Battery: The drone battery mainly uses a 6000mA lithium battery.
5. The tracking gimbal for visual dynamic capture of unmanned aerial vehicles according to claim 1, characterized in that: The design of the drone sensor includes the following steps: Step S1: The main function of the accelerometer is to provide the acceleration force of the drone. The direction of the force is the XYZ three axes, which determines the tilt angle of the drone when it is stationary. The accelerometer is also used to provide horizontal and vertical linear acceleration, which can be used to monitor the vibration of the drone. Step S2: The gyroscope monitors the angular velocity of the XYZ axes and calculates the change rate of the pitch, roll and yaw angles. The change of the angle information obtained by the gyroscope can be used to maintain the stability of the drone and prevent shaking. The gyroscope can also ensure that the drone rotates according to the angle set by the user control device; Step S3: The function of the magnetic compass is to provide a sense of direction for the drone. It can provide data on the magnetic field to which the device is subjected in each axis of XYZ. Then the algorithm of the microcontroller will process the relevant magnetic field data to provide the heading angle related to the magnetic north pole. The final detection of the geographic orientation is achieved by relying on these data. Step S4: In the application of drones, the magnetic compass is not only used to sense the direction, but also can be used to detect the surrounding magnetism and ferrous metals, such as electrodes, wires, vehicles, and other drones, to avoid flight accidents; Step S5: The barometer is used to calculate the current altitude information through the atmospheric pressure. The pressure sensor can detect the atmospheric pressure of the earth. The data provided by the barometer can assist the drone in navigation and flying to a predetermined altitude. Step S6: GPS provides the user with the specific location information of the drone, and the location information of the drone can be monitored through signals even if it is out of the field of view.
6. The tracking gimbal for visual dynamic capture of an unmanned aerial vehicle according to claim 1, characterized in that: The steps of designing the drone tracking gimbal are as follows: Step S1: Install Cygwin, which is a UNIX-like simulation environment running on the Windows platform, and add CYGWIN_HOME to the system environment variables; Step S2: Install MAVProxy, install the executable file and accept the license and all default installation options; Step S3: Run SITL, open a Cygwin64 terminal, run SITL to start the simulation program, then SITL and MAVProxy will be opened, and MAVProxy will open three windows: the first one can input commands directly to SITL, the second one can display real-time flight control status and information, and the third one is a map showing the current location of the drone, and you can control the movement of the drone and perform flight missions through commands / right-clicking the mouse; Step S4: Send instructions to SITL to control the drone. Enter the command "modeguided;armthrottle;takeoff40;" here, the drone will switch to the guided mode, increase the throttle, and take off to a height of 40 meters. Then enter the following command rc31500;modecircle;paramsetcircle_radius2000;", the drone will start the circle mode and set the flight radius to 2000 cm or 20 meters. If you want the drone to return successfully, just enter the command modertl: 1) Download dronekit-python, compile and install dronckit, and install python2.7; 2) Run SITL and MAVProxy; 3) Install MissionPlanner, which is the drone ground station and can also be used to simulate the flight control connection, here through UDP connection; Step S5: Run the flight control script. First, use UDP to connect to the SITL simulator through the local port 14450: Then perform a pre-takeoff check on the drone and unlock the drone: Next, switch the drone flight mode to "GUIDED" mode, confirm the motor unlocking status, send a takeoff command, the drone takes off and flies to the target altitude, and continue to send commands to let the drone fly to the first set waypoint. The waypoint can be set to a certain longitude and latitude through the parameter LocationGlobalRelative, and then send a return command. The drone will automatically return to the top of the home point and then automatically land to complete the flight mission. Finally, the output of the flight control module is the control command. The next step is to transmit the control command to the flight control, and then control the drone to autonomously track the target according to the calculation results in OFFBOARD mode.
Citation Information
Patent Citations
Target detection tracking system for miniature double-axis visual stable holder
CN108107920A
Modular multifunctional airborne computer system
CN112198903A
Four-rotor unmanned aerial vehicle target tracking method based on active visual perception
CN113674310A
Stable control system who trails moving target of rotor unmanned aerial vehicle based on vision
CN206532142U
