A visually synchronized unmanned aerial vehicle flight control system and control method

Abstract

The present invention proposes a visually synchronized UAV flight control system and control method. The analog signal of the UAV sensor is collected, and the analog signal is converted into a UAV digital signal through analog-to-digital conversion; the obtained UAV digital signal is transmitted to the control terminal, and the UAV visual data is obtained after processing; the UAV data collection threshold is set, the monitoring control threshold is monitored, and the warning information is obtained when the threshold is exceeded; the control terminal sends the UAV control command, the control command is used for neural network training, and the UAV control neural network is obtained through deep learning, thereby improving the precise control of the UAV, reducing the impact of the environment on the UAV, increasing the field of view of the UAV, and increasing the practicality and safety of the UAV in complex environments.

Description

Technical Field

[0001] The invention proposes a vision-synchronized unmanned aerial vehicle flight control system and a control method, belonging to the technical field of unmanned aerial vehicle intelligent control. Background Art

[0002] Currently, drone technology is in a state of rapid development, but the control of drones is not precise enough, and they are easily affected by wind and unstable. In addition, the drone's field of view has blind spots, and only part of the drone's angle can be viewed through the control screen, making it difficult to enter complex environments. At the same time, the drone's battery power limits its working time, resulting in the drone's inability to return normally and causing the drone to be lost. Summary of the invention

[0003] The present invention provides a vision-synchronized UAV flight control system and control method, which are used to improve the precise control of the UAV, reduce the impact of the environment on the UAV, increase the field of view of the UAV, and increase the practicality and safety of the UAV in complex environments.

[0004] The present invention proposes a visual synchronization unmanned aerial vehicle flight control method, the method comprising:

[0005] Collect the analog signal of the drone sensor, and obtain the drone digital signal through analog-to-digital conversion;

[0006] The acquired UAV digital signal is transmitted to the control terminal, and the UAV visualization data is obtained after processing;

[0007] Set the drone data collection threshold, monitoring control threshold, and obtain warning information when the threshold is exceeded;

[0008] The control end sends the drone control instructions, the control instructions are used to train the neural network, and deep learning is used to obtain the drone control neural network.

[0009] Furthermore, the acquisition of the drone sensor analog signal, and the acquisition of the drone digital signal by analog-to-digital conversion of the analog signal, includes:

[0010] Combined video sensors, with overlapping sensor viewing angles, can obtain multiple sets of drone visual signals;

[0011] Locate the drone through the position sensor and obtain the drone's spatial position signal;

[0012] Detect the real-time power of the drone through power detection and obtain the remaining power signal of the drone;

[0013] The wind direction, wind force and wind speed of the current drone are measured through the wind sensor to obtain the drone's wind signal;

[0014] The signal acquired by the drone is converted into digital form through an analog-to-digital converter to obtain the digital signal of the drone.

[0015] Furthermore, the acquired drone digital signal is transmitted to the control terminal, and the drone visualization data is obtained after processing, including:

[0016] The drone digital signal is transmitted to the control end through the WiFi wireless network, and the digital signal is packaged and distributed at the control end, and the corresponding processor obtains the digital signal to be processed;

[0017] Extract the spatial position signal of the UAV, establish a spatial coordinate system with the position of the control end as the coordinate origin, and obtain the UAV digital processing coordinate system;

[0018] Combine the video sensor perspectives to locate spatial coordinates, intercept the spatial coordinates of the intersection, fuse the intersection data, and obtain the UAV wide-angle video digital signal;

[0019] The UAV digital processing coordinate system is combined with the UAV wide-angle video digital signal to obtain the digital signal of the distance between the UAV and the target point through spatial coordinate calculation;

[0020] The processed drone digital signal is converted into digital-to-analog through a digital-to-analog converter, and the data is projected onto the LCD display screen at the control end to obtain a visual image of the drone's wide-angle synchronized video, real-time power, wind speed and direction, and the distance between the drone and the target point.

[0021] Furthermore, the setting of the drone data collection threshold, the monitoring control threshold, and obtaining warning information when the threshold is exceeded includes:

[0022] Set the distance threshold between the drone and the controller. If the distance exceeds the threshold, a prompt will be given to obtain the control distance warning information.

[0023] Set the contact distance threshold between the drone and the target, monitor the distance between the drone and the target in real time, and obtain contact distance warning information when the contact distance threshold is reached;

[0024] Set the wind speed threshold of the drone. When the actual wind speed exceeds the torque range of the drone rotor, issue a wind danger control warning.

[0025] Set the drone power threshold to obtain the current drone return power requirement. If the drone return power exceeds the limit, a return power warning will be issued.

[0026] Furthermore, the control end sends a UAV control instruction, the control instruction performs neural network training, and deep learning obtains the UAV control neural network, including:

[0027] When affected by wind, the wind factor neural network is preset for training, and deep learning of wind utilization is performed to obtain a wind optimization control neural network;

[0028] When the battery level is affected, the preset neural network training is performed on the remaining battery level of the drone, and deep learning is used to obtain the remaining battery level return control neural network;

[0029] When the drone is affected by close contact, neural network training is performed on the close-range braking of the drone, and deep learning is used to obtain the contact distance braking control neural network;

[0030] When affected by the signal transmission threshold, neural network training is performed on the edge control distance braking of the drone, and deep learning is used to obtain the control distance braking control neural network.

[0031] A visually synchronized UAV flight control system, the system comprising:

[0032] The data acquisition module is used to collect the analog signals of the drone sensors and convert the analog signals into digital signals to obtain the digital signals of the drone;

[0033] The transmission processing module is used to transmit the UAV digital signal to the control terminal and obtain the UAV visualization data after processing;

[0034] Data warning module, used to set the threshold of drone data collection, monitor and control the threshold, and obtain warning information when the threshold is exceeded;

[0035] The learning control module is used to receive the drone control instructions sent by the control end, perform neural network training on the control instructions, and obtain the drone control neural network through deep learning.

[0036] Furthermore, the data acquisition module includes:

[0037] Video data module, used to combine video sensors, collect video data and obtain multiple sets of drone visual signals;

[0038] Position data module, used to locate the drone and obtain the drone's spatial position signal;

[0039] The power detection module is used to detect the real-time power of the drone and obtain the remaining power signal of the drone;

[0040] The wind force detection module is used to measure the wind direction, wind force and wind speed of the current drone and obtain the drone wind force signal;

[0041] The signal conversion module performs analog-to-digital conversion on the signal acquired by the drone to obtain the drone digital signal.

[0042] Furthermore, the transmission processing module includes:

[0043] The transmission and distribution module is used to transmit the UAV digital signal to the control end, package and distribute the digital signal at the control end, and send the digital signal to be processed to the corresponding processor;

[0044] The position processing module is used to extract the spatial position signal of the UAV, establish a spatial coordinate system with the position of the control end as the coordinate origin, and obtain the UAV digital processing coordinate system;

[0045] The video processing module is used to combine the video sensor's viewing angle to locate the spatial coordinates, intercept the spatial coordinates of the intersection, fuse the intersection data, and obtain the UAV wide-angle video digital signal;

[0046] The distance processing module is used to combine the UAV digital processing coordinate system with the UAV wide-angle video digital signal, and obtain the digital signal of the distance between the UAV and the target point through spatial coordinate calculation;

[0047] The visual conversion module is used to convert the processed drone digital signal into digital-to-analog format through a digital-to-analog converter, and the data is displayed on the LCD display screen at the control end to obtain a visual image of the drone's wide-angle synchronized video, real-time power, wind speed and direction, and the distance between the drone and the target point.

[0048] Furthermore, the data early warning module includes:

[0049] The control distance warning module is used to set the distance threshold between the drone and the control terminal, issue a prompt when the distance threshold is exceeded, and obtain control distance warning information;

[0050] The contact distance warning module is used to set the contact distance threshold between the drone and the target, monitor the distance between the drone and the target in real time, and obtain contact distance warning information when the contact distance threshold is reached;

[0051] The wind out-of-control warning module is used to set the wind threshold of the drone. When the actual wind speed reaches the torque range of the drone rotor, a wind danger control warning is issued;

[0052] The return power warning module is used to set the drone power threshold, obtain the current drone return power required, and issue a return power warning when the drone return power is exceeded.

[0053] Furthermore, the learning control module includes:

[0054] The wind power utilization control module is used to preset wind power factors for neural network training, conduct deep learning on wind power utilization, and obtain a wind power optimization control neural network;

[0055] The battery return control module is used to perform preset neural network training on the remaining battery of the drone, and obtain the remaining battery return control neural network through deep learning;

[0056] The contact distance control module is used to train the neural network for close-range braking of the drone and obtain the contact distance braking control neural network through deep learning;

[0057] The edge distance control module is used to train the neural network for the edge control distance braking of the drone, and to obtain the control distance braking control neural network through deep learning.

[0058] Beneficial effects of the present invention:

[0059] By collecting drone signals and obtaining multiple sets of cross-video data, combined with the spatial coordinate system generated by the position sensor, the distance between the drone and the target and the control end can be obtained. At the same time, by obtaining wind data, the drone can add wind factors according to the control instructions to obtain the actual power consumption and current speed of the drone, thereby controlling the drone motor for speed control and completing the stable control of the drone under wind factors. In addition, adding power factors can obtain the latest drone power consumption, and obtain return data through predictive calculation. At the same time, the signals collected by the drone are transmitted after analog-to-digital conversion, which can speed up the transmission speed of the drone signal and facilitate the processing of digital signals and the optimization of drone data, thereby facilitating the setting of thresholds for drones and avoiding losses caused by drones. At the same time, neural network training is performed on the drone's commands to speed up the execution response speed and accuracy of the drone's commands. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 is a flow chart of the method of the present invention;

[0061] Figure 2 The system framework diagram of the system described in the present invention. DETAILED DESCRIPTION

[0062] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0063] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. The embodiments described are only a part of the embodiments of the present invention, rather than 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.

[0064] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art of the present invention. The terms used in the specification of the present invention herein are only for the purpose of describing specific embodiments and are not intended to limit the present invention.

[0065] One embodiment of the present invention provides a visually synchronized UAV flight control method, the method comprising:

[0066] Collect the analog signal of the drone sensor, and obtain the drone digital signal through analog-to-digital conversion;

[0067] The acquired UAV digital signal is transmitted to the control terminal, and the UAV visualization data is obtained after processing;

[0068] Set the drone data collection threshold, monitoring control threshold, and obtain warning information when the threshold is exceeded;

[0069] The control end sends the drone control instructions, the control instructions are used to train the neural network, and deep learning is used to obtain the drone control neural network.

[0070] The working principle of the above technical solution is:

[0071] First, the drone analog signal is collected by the signal acquisition device on the drone, and the analog signal is converted into a digital signal by an analog-to-digital conversion device;

[0072] Then, the acquired drone digital signal is transmitted to the control terminal through the wifi device, and the digital signal is processed and converted into drone visualization data;

[0073] After that, set the drone data collection threshold and monitoring control threshold, and obtain warning information when the drone data exceeds the threshold;

[0074] Finally, the control end sends control instructions to the drone through the data transmitted back by the drone, and the control instructions are used to train the neural network, and deep learning is used to obtain the drone control neural network.

[0075] The effects of the above technical solution are:

[0076] The signal acquisition device on the drone realizes the real-time acquisition of analog signals. The analog signals are converted into digital signals through the analog-to-digital conversion device, which is convenient for subsequent processing and transmission. The digital signals collected by the drone are transmitted to the control terminal through the WiFi device. The digital signals are processed and converted into drone visualization data, which enables the control terminal to obtain the status and environmental information of the drone more intuitively.

[0077] By setting the thresholds for drone data collection in the technical solution and monitoring these thresholds in real time, when the data collected by the drone exceeds the preset threshold, the system can issue a warning message in time to remind the control end to make corresponding adjustments to the drone or take countermeasures to avoid losses caused by the drone;

[0078] By training and learning UAV data through neural networks, the control instructions of UAVs can be continuously optimized. Through deep learning algorithms, UAVs can better adapt to various environments and mission requirements, improve their autonomous control capabilities and execution efficiency, and help improve the stability and adaptability of UAVs, enabling them to complete tasks more reliably and efficiently in complex environments, and improve the flexibility and safety of operations.

[0079] In one embodiment of the present invention, an analog signal of a drone sensor is collected, and a digital signal of the drone is obtained by analog-to-digital conversion of the analog signal, including:

[0080] Combined video sensors, with overlapping sensor viewing angles, can obtain multiple sets of drone visual signals;

[0081] Locate the drone through the position sensor and obtain the drone's spatial position signal;

[0082] Detect the real-time power of the drone through power detection and obtain the remaining power signal of the drone;

[0083] The wind direction, wind force and wind speed of the current drone are measured through the wind sensor to obtain the drone's wind signal;

[0084] The signal acquired by the drone is converted into digital form through an analog-to-digital converter to obtain the digital signal of the drone.

[0085] The working principle of the above technical solution is:

[0086] First, the video sensors are combined, and the edges of the sensor viewing angles overlap to obtain multiple sets of drone visual signals;

[0087] Then, the drone is located through the position sensor to obtain the drone's spatial position signal;

[0088] Afterwards, the real-time power of the drone is detected through power detection to obtain the remaining power signal of the drone;

[0089] Secondly, the wind direction, wind force and wind speed of the current UAV are measured through the wind sensor to obtain the UAV wind signal;

[0090] Finally, the signal acquired by the drone is converted into digital form through an analog-to-digital converter to obtain the digital signal of the drone.

[0091] The effects of the above technical solution are:

[0092] By combining multiple video sensors and overlapping the edges of the sensor's viewing angle, multiple sets of UAV visual signals can be obtained, which helps to expand the UAV's field of view, gain a more comprehensive understanding of the UAV's surroundings, and improve the accuracy and stability of visual perception;

[0093] Through the position sensor, the drone can obtain accurate spatial position signals, which enables the drone to understand its own position information in real time and provide basic data support for navigation, path planning and distance measurement;

[0094] Through the power detection device, the drone can obtain the remaining power signal in real time, which helps the drone to reasonably allocate energy when performing tasks, plan the drone mission in advance, and avoid running out of power and being unable to return;

[0095] The wind direction and wind force of the drone are measured by the wind sensor, so as to obtain the wind force signal of the drone, which provides basic data support for the stable flight, attitude control and wind force of the drone.

[0096] The analog signal obtained by the drone is converted into a digital signal through an analog-to-digital converter, because digital signals are easy to transmit, process and analyze, which improves the efficiency and accuracy of signal processing;

[0097] By acquiring multiple types of signals, drones can perform comprehensive information processing and optimization, which helps improve the decision-making ability of the drone control end, enabling it to perform tasks more intelligently in complex environments, thereby improving the overall performance and application range of the drone.

[0098] In one embodiment of the present invention, the acquired drone digital signal is transmitted to a control terminal, and the drone visualization data is obtained after processing, including:

[0099] The drone digital signal is transmitted to the control end through the WiFi wireless network, and the digital signal is packaged and distributed at the control end, and the corresponding processor obtains the digital signal to be processed;

[0100] Extract the spatial position signal of the UAV, establish a spatial coordinate system with the position of the control end as the coordinate origin, and obtain the UAV digital processing coordinate system;

[0101] Combine the video sensor perspectives to locate spatial coordinates, intercept the spatial coordinates of the intersection, fuse the intersection data, and obtain the UAV wide-angle video digital signal;

[0102] The UAV digital processing coordinate system is combined with the UAV wide-angle video digital signal, and the digital signal of the distance between the UAV and the target point is obtained through spatial coordinate calculation; the processed UAV digital signal is converted into digital-to-analog through a digital-to-analog converter, and the data is projected onto the LCD display screen at the control end to obtain the UAV wide-angle synchronous video, real-time power, wind speed and direction, and a visual image of the distance between the UAV and the target point.

[0103] The working principle of the above technical solution is:

[0104] First, the drone digital signal is transmitted to the control end through the WiFi wireless network, where the digital signal is packaged and distributed, and the corresponding processor obtains the digital signal to be processed;

[0105] Then, the spatial position signal of the UAV is extracted, a spatial coordinate system is established with the position of the control end as the coordinate origin, and the digital processing coordinate system of the UAV is obtained;

[0106] Secondly, the spatial coordinate positioning is performed by combining the video sensor perspectives, intercepting the spatial coordinates of the intersection part, fusing the intersecting data, and obtaining the wide-angle video digital signal of the drone; the spatial coordinate positioning of the video sensor perspective is performed, and when the videos collected by different video sensors have the same spatial coordinates, the image data with the same spatial coordinates are merged, and the image curves with overlapping coordinates are smoothly adjusted;

[0107] Afterwards, the drone digital processing coordinate system is combined with the drone wide-angle video digital signal, and the digital signal of the drone's distance from the target point is obtained through spatial coordinate calculation; the drone takes off from the control end, and the length, width and height data of the drone's current distance from the origin of the control end are obtained through the position sensor. Assuming the drone's spatial coordinates are (x, y, z), the Manhattan distance of the drone flying back to the control end is L, L = |x| + |y| + |z|,

[0108] The straight-line distance between the drone and the origin of the control end is r.

[0109] Put the detection target into the spatial coordinate system, the center point of the monitoring target is (x1, y1, z1), the distance between the drone and the target is L'=|x-x1|+|y-y1|+|z-z1|, when L'=0, the drone contacts and collides with the target point;

[0110] Finally, the processed drone digital signal is converted into digital-to-analog through a digital-to-analog converter, and the data is projected onto the LCD display screen at the control end to obtain a visual image of the drone's wide-angle synchronized video, real-time power, wind speed and direction, and the distance between the drone and the target point.

[0111] The effects of the above technical solution are:

[0112] The drone digital signal is transmitted to the control end through the WiFi wireless network, and the control end packages and distributes these digital signals to ensure the effective transmission and real-time processing of the drone digital signal;

[0113] A spatial coordinate system is established with the position of the drone or the control terminal as the coordinate origin, and the spatial position signal of the drone is extracted and located, which provides a geographic reference framework for subsequent data processing and decision-making, and facilitates drone spatial analysis;

[0114] By combining the perspectives of multiple video sensors and intercepting the spatial coordinates of the intersection, the intersecting data is fused to enhance the consistency and accuracy of data processing and provide a more comprehensive visual environment perception.

[0115] Combining the UAV digital processing coordinate system with the wide-angle video digital signal, the distance information between the UAV and the target point can be obtained in real time through spatial coordinate calculation, providing key data for the positioning, control and mission execution of the UAV.

[0116] The processed drone digital signal is projected onto the LCD display screen at the control end, generating visual images such as wide-angle synchronous video, real-time power, wind speed and direction, and the distance between the drone and the target point, which can help the control end to intuitively understand the status of the drone and the surrounding environment, so as to make quick and accurate decisions;

[0117] Through this series of technical effects, drones can be positioned, controlled and perform tasks in complex environments, while providing real-time status and environmental information to the control end, making the entire system more intelligent, efficient and reliable.

[0118] In one embodiment of the present invention, a threshold for collecting data from a drone is set, a monitoring and control threshold is monitored, and warning information is obtained when the threshold is exceeded, including:

[0119] Set the distance threshold between the drone and the controller. If the distance exceeds the threshold, a prompt will be given to obtain the control distance warning information.

[0120] Set the contact distance threshold between the drone and the target, monitor the distance between the drone and the target in real time, and obtain contact distance warning information when the contact distance threshold is reached;

[0121] Set the wind speed threshold of the drone. When the actual wind speed exceeds the torque range of the drone rotor, issue a wind danger control warning.

[0122] Set the drone power threshold to obtain the current drone return power requirement. If the drone return power exceeds the limit, a return power warning will be issued.

[0123] The working principle of the above technical solution is:

[0124] Assume the weight of the drone is m, the vertical torque of the drone is F1, the working power corresponding to torque F1 is P1, the horizontal longitudinal torque is F2, the working power corresponding to torque F2 is P2, the horizontal lateral torque is F3, the working power corresponding to torque F2 is P3, the real-time wind force is F4, and the wind speed v 风 The angle between the wind force F4 and the vertical upward direction of the UAV is w1, the angle between F4 and the horizontal longitudinal outward direction of the UAV is w2, and the angle between F4 and the horizontal lateral outward direction of the UAV is w3. The UAV operation time is t. The vertical force F of the wind force F4 dispersed to the UAV is 风垂 =F4*cosw1, the wind force F4 dispersed to the horizontal longitudinal force F of the drone 风纵 =F4*cosw2, F3 is the force F distributed to the horizontal longitudinal direction of the drone 风横 =F4*cosw3,

[0125] In a windy environment, the vertical force F on the drone 垂 =F 风垂 +mg, g is the acceleration due to gravity, when F1=F 垂 When F1≠F 垂 When the UAV moves in the vertical direction with an acceleration a 垂 To speed up,

[0126] In a windy environment, the UAV is subjected to horizontal longitudinal force F 纵 =F 风纵 +F2, when F 纵 = 0, the drone maintains horizontal longitudinal braking. 纵 ≠0, the UAV moves in the horizontal longitudinal direction with an acceleration a 纵 To speed up,

[0127] In a windy environment, the horizontal lateral force F of the drone is 横 =F 风横 +F3, when F 横 = 0, the drone maintains horizontal lateral braking. 横 ≠0, the UAV moves horizontally with an acceleration a 横 To speed up,

[0128] The time for the drone to accelerate is t 加速 , uniform time is t 匀 , braking time is t 制 , the target point coordinates are (x1, y1, z1), the acceleration time distance, the speed of the drone before acceleration is v 0 , after acceleration, the speed is v 末 ,

[0129]

[0130]

[0131]

[0132] Power consumption during acceleration

[0133] Power consumption during constant speed

[0134] The emergency braking time of the drone is t 制 After braking, the aircraft is stationary in the air, and the braking energy

[0135]

[0136] Vertical braking distance

[0137] Horizontal longitudinal braking distance

[0138] Horizontal braking distance

[0139] When the drone is stationary, the vertical torque is the opposite force of the aircraft's gravity and the vertical wind force, the horizontal longitudinal torque is the opposite force of the wind's horizontal longitudinal force, and the horizontal lateral torque is the opposite force of the wind's horizontal lateral force.

[0140]

[0141] First, set the distance threshold between the drone and the control terminal, and give a prompt when the distance exceeds the threshold to obtain control distance warning information;

[0142] The straight-line distance between the drone and the origin of the control end is r. When r≥200 meters, the control is too far and a warning message is issued;

[0143] Then, the contact distance threshold between the drone and the target is set, and the distance between the drone and the target is monitored in real time. When the contact distance threshold is reached, the contact distance warning information is obtained; the Manhattan distance of the drone target in each direction is compared with the braking distance. When the distance difference in the three directions is close to 0, that is, |x-x1|-S 垂制 →0, and |y-y1|-S 纵制 →0, and |z-z1|-S 横制 →0, conduct contact alarm warning;

[0144] After that, set the wind speed threshold of the drone. When the actual wind speed exceeds the rotor torque range of the drone, that is, when F1-F 垂 ≤0.3F1, or F2-F纵 ≤0.3F2, or F3-F 横 ≤0.3F3, a wind danger control warning is issued; when F1≤F 垂 , or F2≤F 纵 , or F3≤F 横 When the wind is out of control, a flight ban warning will be issued;

[0145] Finally, set the drone power threshold to obtain the current drone return power required. If the drone return power exceeds the return power, a return power warning will be issued, and the remaining power of the drone is Q 剩 =Q 总 -Q 加速 -Q 匀速 -Q 制动 -Q 静止 ,

[0146] Remaining flight time of the drone

[0147] Remaining driving distance

[0148] Compare the Manhattan distance L between the drone and the controller with the remaining driving distance.

[0149] If S 剩 -L≤0.2S 剩 , then a return power warning is issued. If S 剩 ≤L, then look for a safe place to land.

[0150] The effects of the above technical solution are:

[0151] By setting the distance threshold between the drone and the control terminal, when the distance exceeds the threshold, the system will issue a warning message to ensure that the communication between the control terminal and the drone will not be interrupted due to the long distance, thus improving the stability and reliability of communication;

[0152] By setting a distance threshold for the drone to contact the target in the technical solution, when the drone approaches the target to a certain distance, the system will issue a warning message to remind the controller to pay attention to the operation of the drone to avoid collision or unnecessary contact;

[0153] By setting the wind speed threshold of the drone, when the actual wind speed exceeds the torque range of the drone rotor, the system will issue a wind danger control warning, which helps prevent damage or loss of control of the drone due to excessive wind speed and improves the safety of the drone;

[0154] In addition, the power threshold of the drone is also taken into consideration. When the power required for the drone to return exceeds the preset threshold, the system will issue a return power warning, which will help the drone return in time or find a parking spot when the power is low, ensuring the continuity and safety of the mission.

[0155] Through a series of early warning measures, the overall safety of drones has been significantly improved. The control end can take corresponding measures in a timely manner based on the early warning information to avoid potential risks and accidents, further enhancing the reliability and practicality of drones.

[0156] In one embodiment of the present invention, the control end sends a drone control instruction, the control instruction performs neural network training, and deep learning obtains the drone control neural network, including:

[0157] When affected by wind, the wind factor neural network is preset for training, and deep learning of wind utilization is performed to obtain a wind optimization control neural network;

[0158] When the battery level is affected, the preset neural network training is performed on the remaining battery level of the drone, and deep learning is used to obtain the remaining battery level return control neural network;

[0159] When the drone is affected by close contact, neural network training is performed on the close-range braking of the drone, and deep learning is used to obtain the contact distance braking control neural network;

[0160] When affected by the signal transmission threshold, neural network training is performed on the edge control distance braking of the drone, and deep learning is used to obtain the control distance braking control neural network.

[0161] The working principle of the above technical solution is:

[0162] First, when the drone is affected by wind, the preset wind factor neural network is trained to regulate the torque of the drone. Through force analysis and speed calculation, the wind force is deeply learned in the downwind and headwind control of the drone, making full use of the impact of wind on the drone, speeding up the drone's reaction speed, and obtaining the wind optimization control neural network.

[0163] Secondly, monitor the power consumption of the drone, perform preset neural network training on the remaining power of the drone, calculate the power consumption of the drone when controlling the drone, calculate the power consumption of the drone in different environments and different actions, and obtain the remaining power return control neural network through deep learning integrated calculation;

[0164] Afterwards, when the drone is affected by a close contact with a target, a neural network is trained for close-range braking of the drone, and the braking time and distance of the drone are gradually reduced while ensuring the stability of the drone. Deep learning is used to obtain the contact distance braking control neural network;

[0165] Finally, when the drone is affected by the signal transmission threshold, the distance and the control delay of the drone are gradually counted. When the operation delay is close to 0.3 seconds, it is the obvious human reaction time. The distance between the drone and the control end is shortened, so as to train the neural network for the edge control distance braking of the drone, and obtain the control distance braking control neural network through deep learning.

[0166] The effects of the above technical solution are:

[0167] By presetting the wind factor neural network and performing deep learning, the drone can better utilize wind power for flight, which can improve the drone's energy efficiency and stability in strong wind conditions;

[0168] By monitoring the power consumption of the drone and pre-training the neural network on the remaining power, through deep learning, the drone can more intelligently plan its return or find a safe parking spot to ensure the safe completion of the mission;

[0169] By monitoring the distance of the drone to the target and optimizing the close-range braking control through neural network training, the drone can approach the target more accurately when performing tasks, while avoiding collision or unnecessary contact;

[0170] By monitoring the UAV signal transmission distance threshold and conducting braking training and deep learning on the edge control distance of the UAV, it helps the UAV maintain stable communication with the control end in complex environments, ensures the reliability and stability of remote control, and avoids losses caused by loss of control of the UAV; through a series of neural network training and deep learning, the UAV provides strong support for the intelligent control of the UAV and enhances its adaptability and performance in various environments and mission conditions.

[0171] One embodiment of the present invention is a visually synchronized UAV flight control system, the system comprising:

[0172] The data acquisition module is used to collect the analog signals of the drone sensors and convert the analog signals into digital signals to obtain the digital signals of the drone;

[0173] The transmission processing module is used to transmit the UAV digital signal to the control terminal and obtain the UAV visualization data after processing;

[0174] Data warning module, used to set the threshold of drone data collection, monitor and control the threshold, and obtain warning information when the threshold is exceeded;

[0175] The learning control module is used to receive the drone control instructions sent by the control end, perform neural network training on the control instructions, and obtain the drone control neural network through deep learning.

[0176] The working principle of the above technical solution is:

[0177] First, the UAV analog signal is collected through the data acquisition module and converted into a digital signal;

[0178] Then, the acquired UAV digital signal is transmitted to the control terminal through the transmission processing module, and the digital signal is converted into UAV visualization data;

[0179] After that, the drone data collection threshold is set through the data warning module, and the monitoring and control threshold is set. When the drone data exceeds the threshold, the warning information is obtained;

[0180] Finally, the data sent back by the drone is analyzed, and control instructions are sent to the drone through the learning control module. The control instructions are used to train the neural network, and deep learning is used to obtain the drone control neural network.

[0181] The effects of the above technical solution are:

[0182] The data acquisition module collects signals from the drone, realizing real-time acquisition of analog signals, which are converted into digital signals to facilitate subsequent signal processing and transmission;

[0183] The digital signals collected by the drone are transmitted to the control terminal through the transmission processing module, and then the digital signals are processed and converted into visual data, which enables the control terminal to obtain the status and environmental information of the drone more intuitively;

[0184] The data warning module sets the thresholds for drone data collection and monitors these thresholds in real time. When the data collected by the drone exceeds the preset threshold, the system can issue a warning message in time to remind the control end to make corresponding adjustments to the drone or take countermeasures to avoid losses caused by the drone.

[0185] By using the learning control module to train and learn the UAV data, the neural network can continuously optimize the UAV's control instructions. Through the deep learning algorithm, the UAV can better adapt to various environments and mission requirements, improve its autonomous control capabilities and execution efficiency, and help improve the stability and adaptability of the UAV, enabling it to complete tasks more reliably and efficiently in complex environments, and improve the flexibility and safety of operations.

[0186] In one embodiment of the present invention, the data acquisition module includes:

[0187] Video data module, used to combine video sensors, collect video data and obtain multiple sets of drone visual signals;

[0188] Position data module, used to locate the drone and obtain the drone's spatial position signal;

[0189] The power detection module is used to detect the real-time power of the drone and obtain the remaining power signal of the drone;

[0190] The wind force detection module is used to measure the wind direction, wind force and wind speed of the current drone and obtain the drone wind force signal;

[0191] The signal conversion module performs analog-to-digital conversion on the signal acquired by the drone to obtain the drone digital signal.

[0192] The working principle of the above technical solution is:

[0193] First, the video sensors are combined through the video data module, multiple video sensors are overlapped with the sensor viewing angle edges, and multiple sets of drone visual signals are obtained;

[0194] Then, the position of the drone is located through the position data module to obtain the drone's spatial position signal;

[0195] Afterwards, the real-time power of the drone is detected through the power detection module to obtain the remaining power signal of the drone;

[0196] Secondly, the wind detection module is used to measure the wind direction, wind force and wind speed of the current drone to obtain the drone wind signal;

[0197] Finally, the acquired signal is converted into digital form through the signal conversion module to obtain the drone digital signal.

[0198] The effects of the above technical solution are:

[0199] By combining multiple video sensors and overlapping the sensor's viewing angle edges through the video data module, multiple sets of UAV visual signals can be obtained, which helps to expand the UAV's field of view, gain a more comprehensive understanding of the UAV's surroundings, and improve the accuracy and stability of visual perception;

[0200] Through the location data module, the drone can obtain accurate spatial location signals, which enables the drone to understand its own location information in real time and provide basic data support for navigation, path planning and distance measurement;

[0201] Through the power detection module, the drone can obtain the remaining power signal in real time, which helps the drone to reasonably allocate energy when performing tasks, plan drone tasks in advance, and avoid running out of power and being unable to return;

[0202] The wind detection module measures the wind direction, wind force and wind speed of the drone, thereby obtaining the wind signal of the drone, which provides basic data support for the drone's stable flight, attitude control and wind assistance.

[0203] The analog signal obtained by the drone is converted into a digital signal through the signal conversion module, because the digital signal is easy to transmit, process and analyze, which improves the efficiency and accuracy of signal processing;

[0204] By acquiring multiple types of signals, drones can perform comprehensive information processing and optimization, which helps improve the decision-making ability of the drone control end, enabling it to perform tasks more intelligently in complex environments, thereby improving the overall performance and application range of the drone.

[0205] In one embodiment of the present invention, a transmission processing module includes:

[0206] The transmission and distribution module is used to transmit the UAV digital signal to the control end, package and distribute the digital signal at the control end, and send the digital signal to be processed to the corresponding processor;

[0207] The position processing module is used to extract the spatial position signal of the UAV, establish a spatial coordinate system with the position of the control end as the coordinate origin, and obtain the UAV digital processing coordinate system;

[0208] The video processing module is used to combine the video sensor's viewing angle to locate the spatial coordinates, intercept the spatial coordinates of the intersection, fuse the intersection data, and obtain the UAV wide-angle video digital signal;

[0209] The distance processing module is used to combine the UAV digital processing coordinate system with the UAV wide-angle video digital signal, and obtain the digital signal of the distance between the UAV and the target point through spatial coordinate calculation;

[0210] The visual conversion module is used to convert the processed drone digital signal into digital-to-analog format through a digital-to-analog converter, and the data is displayed on the LCD display screen at the control end to obtain a visual image of the drone's wide-angle synchronized video, real-time power, wind speed and direction, and the distance between the drone and the target point.

[0211] The working principle of the above technical solution is:

[0212] First, the UAV digital signal is transmitted to the control end through the transmission and distribution module, and the digital signal is packaged and distributed at the control end, and the corresponding processor obtains the digital signal to be processed;

[0213] Then, the spatial position signal of the UAV is extracted through the position processing module, a spatial coordinate system is established with the position of the control end as the coordinate origin, and the digital processing coordinate system of the UAV is obtained;

[0214] Secondly, through the video processing module, the spatial coordinates of the video perspective are located, the spatial coordinates of the intersection are intercepted, the intersecting data are fused, and the wide-angle video digital signal of the drone is obtained. When the videos collected by different video sensors have the same spatial coordinates, the image data with the same spatial coordinates are merged, and the image curves with overlapping coordinates are smoothly adjusted;

[0215] Afterwards, the drone digital processing coordinate system is combined with the drone wide-angle video digital signal through the distance processing module, and the digital signal of the drone's distance to the target point is obtained through spatial coordinate calculation; the drone takes off from the control end, and the length, width and height data of the drone's current distance from the origin of the control end are obtained through the position sensor. Assuming the spatial coordinates of the drone are (x, y, z), the Manhattan distance of the drone flying back to the control end is L, L = |x| + |y| + |z|,

[0216] The straight-line distance between the drone and the origin of the control end is r.

[0217] Put the detection target into the spatial coordinate system, the center point of the monitoring target is (x1, y1, z1), the Manhattan distance between the drone and the target is L'=|x-x1|+|y-y1|+|z-z1|, when L'=0, the drone contacts and collides with the target point;

[0218] Finally, the processed drone digital signal is converted from digital to analog through the visual conversion module, and the data is projected onto the LCD display screen at the control end to obtain a visual image of the drone's wide-angle synchronized video, real-time power, wind speed and direction, and the distance between the drone and the target point.

[0219] The effects of the above technical solution are:

[0220] The UAV digital signals are transmitted to the control end through the transmission and distribution module, and the control end packages and distributes these digital signals, ensuring the effective transmission and real-time processing of the UAV digital signals;

[0221] The position processing module establishes a spatial coordinate system with the position of the drone control terminal as the coordinate origin, and extracts and locates the spatial position signal of the drone, which provides a geographic reference framework for subsequent data processing and decision-making, and facilitates drone spatial analysis;

[0222] The video processing module combines the perspectives of multiple video sensors, intercepts the spatial coordinates of the overlapping parts, and fuses the intersecting data, which enhances the consistency and accuracy of data processing and provides a more comprehensive visual environment perception.

[0223] The distance processing module combines the UAV digital processing coordinate system with the wide-angle video digital signal and calculates the spatial coordinates to obtain the distance information between the UAV and the target point in real time, providing key data for the positioning, control and mission execution of the UAV.

[0224] The digital signal of the drone is projected onto the LCD screen of the control end through the visual conversion module, generating visual images such as wide-angle synchronous video, real-time power, wind speed and direction, and the distance between the drone and the target point, which can help the control end to intuitively understand the status of the drone and the surrounding environment, so as to make quick and accurate decisions;

[0225] Through this series of technical effects, drones can be positioned, controlled and perform tasks in complex environments, while providing real-time status and environmental information to the control end, making the entire control system more intelligent, efficient and reliable.

[0226] In one embodiment of the present invention, a data early warning module includes:

[0227] The control distance warning module is used to set the distance threshold between the drone and the control terminal, issue a prompt when the distance threshold is exceeded, and obtain control distance warning information;

[0228] The contact distance warning module is used to set the contact distance threshold between the drone and the target, monitor the distance between the drone and the target in real time, and obtain contact distance warning information when the contact distance threshold is reached;

[0229] The wind out-of-control warning module is used to set the wind threshold of the drone. When the actual wind speed reaches the torque range of the drone rotor, a wind danger control warning is issued;

[0230] The return power warning module is used to set the drone power threshold, obtain the current drone return power required, and issue a return power warning when the drone return power is exceeded.

[0231] The working principle of the above technical solution is:

[0232] Assume the weight of the drone is m, the vertical torque of the drone is F1, the working power corresponding to torque F1 is P1, the horizontal longitudinal torque is F2, the working power corresponding to torque F2 is P2, the horizontal lateral torque is F3, the working power corresponding to torque F2 is P3, the real-time wind force is F4, and the wind speed v 风 The angle between the wind force F4 and the vertical upward direction of the UAV is w1, the angle between F4 and the horizontal longitudinal outward direction of the UAV is w2, and the angle between F4 and the horizontal lateral outward direction of the UAV is w3. The UAV operation time is t. The vertical force F of the wind force F4 dispersed to the UAV is 风垂 =F4*cosw1, the wind force F4 dispersed to the horizontal longitudinal force F of the drone 风纵 =F4*cosw2, F3 is the force F distributed to the horizontal longitudinal direction of the drone 风横 =F4*cosw3,

[0233] In a windy environment, the vertical force F on the drone 垂 =F 风垂+mg, g is the acceleration due to gravity, when F1=F 垂 When F1≠F 垂 When the UAV moves in the vertical direction with an acceleration a 垂 To speed up,

[0234] In a windy environment, the UAV is subjected to horizontal longitudinal force F 纵 =F 风纵 +F2, when F 纵 = 0, the drone maintains horizontal longitudinal braking. 纵 ≠0, the UAV moves in the horizontal longitudinal direction with an acceleration a 纵 To speed up,

[0235] In a windy environment, the horizontal lateral force F of the drone is 横 =F 风横 +F3, when F 横 = 0, the drone maintains horizontal lateral braking. 横 ≠0, the UAV moves horizontally with an acceleration a 横 To speed up,

[0236] The time for the drone to accelerate is t 加速 , uniform time is t 匀 , braking time is t 制 , the target point coordinates are (x1, y1, z1), the acceleration time distance, the speed of the drone before acceleration is v 0 , after acceleration, the speed is v 末 ,

[0237]

[0238]

[0239]

[0240] Power consumption during acceleration

[0241] Power consumption during constant speed

[0242] The emergency braking time of the drone is t 制 After braking, the aircraft is stationary in the air, and the braking energy

[0243]

[0244] Vertical braking distance

[0245] Horizontal longitudinal braking distance

[0246] Horizontal braking distance

[0247] When the drone is stationary, the vertical torque is the opposite force of the aircraft's gravity and the vertical wind force, the horizontal longitudinal torque is the opposite force of the wind's horizontal longitudinal force, and the horizontal lateral torque is the opposite force of the wind's horizontal lateral force.

[0248]

[0249] First, the distance threshold between the drone and the control end is set through the control distance warning module. The straight-line distance between the drone and the origin of the control end is r. When r≥200 meters, a warning message will be issued to indicate that the control distance is too far, and warning information will be obtained;

[0250] Then, the contact distance warning module is used to set the contact distance threshold between the drone and the target, monitor the distance between the drone and the target in real time, and compare the Manhattan distance of the drone target in each direction with the braking distance. When the distance differences in the three directions are close to 0, that is, |x-x1|-S 垂制 →0, and |y-y1|-S 纵制 →0, and |z-z1|-S 横制 →0, conduct contact alarm warning and obtain contact distance warning information;

[0251] After that, the wind speed threshold of the drone is set through the wind out of control warning module. When the actual wind speed exceeds the torque range of the drone rotor, that is, when F1-F 垂 ≤0.3F1, or F2-F 纵 ≤0.3F2, or F3-F 横 ≤0.3F3, a wind danger control warning is issued; when F1≤F 垂 , or F2≤F 纵 , or F3≤F 横 When the wind is out of control, a flight ban warning will be issued;

[0252] Finally, the return power warning module sets the drone power threshold to obtain the current drone return power required. If the drone return power exceeds the return power, a return power warning is issued, and the remaining power of the drone is Q 剩 =Q 总 -Q 加速 -Q 匀速 -Q 制动 -Q 静止 ,

[0253] Remaining flight time of the drone Remaining driving distance

[0254] Compare the Manhattan distance L between the drone and the controller with the remaining driving distance.

[0255] If S 剩 -L≤0.2S 剩 , then a return power warning is issued. If S 剩 ≤L, then look for a safe place to land.

[0256] The effects of the above technical solution are:

[0257] The distance threshold between the drone and the control terminal is set by controlling the distance warning module. When the distance exceeds the threshold, the system will issue a warning message to ensure that the communication between the control terminal and the drone will not be interrupted due to the long distance, thus improving the stability and reliability of the communication.

[0258] The contact distance warning module sets a distance threshold for the drone to contact the target in the technical solution. When the drone approaches the target to a certain distance, the system will issue a warning message to remind the control end to pay attention to the operation of the drone to avoid collision or unnecessary contact.

[0259] The wind out-of-control warning module can be used to set the wind threshold of the drone. When the actual wind speed exceeds the rotor torque range of the drone, the system will issue a wind danger control warning, which helps prevent damage or loss of control of the drone due to excessive wind and improves the safety of the drone.

[0260] In addition, the drone's power threshold is set through the return power warning module. When the power required for the drone to return exceeds the preset threshold, the system will issue a return power warning, which helps the drone to return in time or find a parking spot when the power is low, ensuring the continuity and safety of the mission.

[0261] In one embodiment of the present invention, a learning control module includes:

[0262] The wind power utilization control module is used to preset wind power factors for neural network training, conduct deep learning on wind power utilization, and obtain a wind power optimization control neural network;

[0263] The battery return control module is used to perform preset neural network training on the remaining battery of the drone, and obtain the remaining battery return control neural network through deep learning;

[0264] The contact distance control module is used to train the neural network for close-range braking of the drone and obtain the contact distance braking control neural network through deep learning;

[0265] The edge distance control module is used to train the neural network for the edge control distance braking of the drone, and to obtain the control distance braking control neural network through deep learning.

[0266] The working principle of the above technical solution is:

[0267] First, when the UAV is affected by wind, the wind factor is preset by the wind control module to train the UAV neural network. By adjusting the torque of the UAV, through force analysis and speed calculation, the wind force is deeply learned in the UAV's downwind control and headwind control, making full use of the impact of wind on the UAV, making the UAV control adapt to the impact of wind, and speeding up the UAV's reaction speed, so as to obtain the wind optimization control neural network;

[0268] Secondly, the power consumption of the drone is monitored through the power return control module, and the preset neural network training is performed on the remaining power of the drone. The power is calculated separately when the drone is controlled, and the power consumption of the drone in different actions under different environments is calculated. The remaining power return control neural network is obtained through deep learning integrated calculation;

[0269] Afterwards, the contact distance control module is used to train the neural network for the close-range braking of the drone. While ensuring the stability of the drone, the controllable braking time and braking distance when the drone is close to the target are gradually reduced, and the contact distance braking control neural network is obtained through deep learning.

[0270] Finally, the edge distance control module is used to monitor the impact of the signal transmission threshold on the drone, and the distance and drone control delay are gradually counted. When the operation delay is close to 0.3 seconds, it is the obvious human reaction time. The operation reaction time must be controlled within 0.3 seconds, thereby shortening the distance between the drone and the control end. The neural network training is performed on the edge control distance braking of the drone, and the signal distance threshold of the drone in different environments is obtained. Deep learning is used to obtain the control distance braking control neural network.

[0271] The effects of the above technical solution are:

[0272] By presetting the wind factor neural network and performing deep learning in the wind power utilization control module, the drone can better utilize wind power for flight, which can improve the drone's energy efficiency and stability in strong wind conditions;

[0273] The power return control module monitors the power consumption of the drone and performs preset neural network training on the remaining power. Through deep learning, the drone can more intelligently plan its return or find a safe parking spot to ensure the safe completion of the mission.

[0274] The contact distance control module monitors the distance of the drone to the target and optimizes the close-range braking control through neural network training, which helps the drone to approach the target more accurately when performing tasks while avoiding collision or unnecessary contact;

[0275] By monitoring the signal transmission distance threshold of the drone through the edge distance control module and conducting braking training and deep learning on the edge control distance of the drone, the drone can maintain stable communication with the control end in complex environments, ensure the reliability and stability of remote control, and avoid losses caused by loss of control of the drone. Through a series of neural network training and deep learning, the drone provides strong support for the intelligent control of the drone and enhances its adaptability and performance in various environments and mission conditions.

[0276] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A visual synchronization UAV flight control method, characterized in that: The method comprises: Collect the analog signal of the drone sensor, and obtain the drone digital signal through analog-to-digital conversion; The acquired UAV digital signal is transmitted to the control terminal, and the UAV visualization data is obtained after processing; Set the drone data collection threshold, monitoring control threshold, and obtain warning information when the threshold is exceeded; The control end sends the drone control command, the control command performs neural network training, and deep learning obtains the drone control neural network; The acquired drone digital signal is transmitted to the control terminal, and the drone visualization data is obtained after processing, including: The drone digital signal is transmitted to the control end through the WiFi wireless network, and the digital signal is packaged and distributed at the control end, and the corresponding processor obtains the digital signal to be processed; Extract the spatial position signal of the UAV, establish a spatial coordinate system with the position of the control end as the coordinate origin, and obtain the UAV digital processing coordinate system; Combine the video sensor perspectives to locate spatial coordinates, intercept the spatial coordinates of the intersection, fuse the intersection data, and obtain the UAV wide-angle video digital signal; The drone digital processing coordinate system is combined with the drone wide-angle video digital signal, and the digital signal of the drone's distance from the target point is obtained through spatial coordinate calculation; the drone takes off from the control end, and the length, width and height data of the drone's current distance from the origin of the control end are obtained through the position sensor. Assuming the drone's spatial coordinates are (x, y, z), the Manhattan distance of the drone flying back to the control end is L, L = |x| + |y| + |z|, The straight-line distance between the drone and the origin of the control end is r. Put the detection target into the spatial coordinate system, the center point of the monitoring target is (x1, y1, z1), the distance between the drone and the target is L'=|x-x1|+|y-y1|+|z-z1|, when L'=0, the drone contacts and collides with the target point; The processed drone digital signal is converted into digital-to-analog through a digital-to-analog converter, and the data is displayed on the LCD display of the control end to obtain the drone's wide-angle synchronous video, real-time power, wind speed and direction, and visual images of the distance between the drone and the target point; The steps of setting the drone data collection threshold, monitoring control threshold, and obtaining warning information when the threshold is exceeded include: Assume the weight of the drone is m, the vertical torque of the drone is F1, the working power corresponding to torque F1 is P1, the horizontal longitudinal torque is F2, the working power corresponding to torque F2 is P2, the horizontal lateral torque is F3, the working power corresponding to torque F2 is P3, the real-time wind force is F4, and the wind speed v 风 The angle between the wind force F4 and the vertical upward direction of the UAV is w1, the angle between F4 and the horizontal longitudinal outward direction of the UAV is w2, and the angle between F4 and the horizontal lateral outward direction of the UAV is w3. The UAV operation time is t. The vertical force F of the wind force F4 dispersed to the UAV is 风垂 =F4*cosw1, the wind force F4 dispersed to the horizontal longitudinal force F of the drone 风纵 =F4*cosw2, F3 is the force F distributed to the horizontal longitudinal direction of the drone 风横 =F4*cosw3, In a windy environment, the vertical force F on the drone 垂 =F 风垂 +mg, g is the acceleration due to gravity, when F1=F 垂 When F1≠F 垂 When the UAV moves in the vertical direction with an acceleration a 垂 To speed up, In a windy environment, the horizontal longitudinal force F of the drone is 纵 =F 风纵 +F2, when F 纵 = 0, the drone maintains horizontal longitudinal braking. 纵 ≠0, the UAV moves in the horizontal longitudinal direction with an acceleration a 纵 To speed up, In a windy environment, the horizontal lateral force F of the drone is 横 =F 风横 +F3, when F 横 = 0, the drone maintains horizontal lateral braking. 横 ≠0, the UAV moves horizontally with an acceleration a 横 To speed up, The time for the drone to accelerate is t 加速 , uniform time is t 匀 , braking time is t 制 , the target point coordinates are (x1, y1, z1), the acceleration time distance, the speed of the drone before acceleration is v0, and the speed after acceleration is v 末 , Power consumption during acceleration Power consumption during constant speed The emergency braking time of the drone is t 制 After braking, the aircraft is stationary in the air, and the braking energy Vertical braking distance Horizontal longitudinal braking distance Horizontal braking distance When the drone is stationary, the vertical torque is the opposite force of the aircraft's gravity and the vertical wind force, the horizontal longitudinal torque is the opposite force of the wind's horizontal longitudinal force, and the horizontal lateral torque is the opposite force of the wind's horizontal lateral force. First, set the distance threshold between the drone and the control terminal, and give a prompt when the distance exceeds the threshold to obtain control distance warning information; The straight-line distance between the drone and the origin of the control end is r. When r≥200 meters, the control is too far and a warning message is issued; Then, the contact distance threshold between the drone and the target is set, and the distance between the drone and the target is monitored in real time. When the contact distance threshold is reached, the contact distance warning information is obtained; the Manhattan distance of the drone target in each direction is compared with the braking distance. When the distance difference in the three directions is close to 0, that is, |x-x1|-S 垂制 →0, and |y-y1|-S 纵制 →0, and |z-z1|-S 横制 →0, conduct contact alarm warning; After that, set the wind speed threshold of the drone. When the actual wind speed exceeds the rotor torque range of the drone, that is, when F1-F 垂 ≤0.3F1, or F2-F 纵 ≤0.3F2, or F3-F 横 ≤0.3F3, a wind danger control warning is issued; when F1≤F 垂 , or F2≤F 纵 , or F3≤F 横 When the wind is out of control, a flight ban warning will be issued; Finally, set the drone power threshold to obtain the current drone return power required. If the drone return power exceeds the return power, a return power warning will be issued, and the remaining power of the drone is Q 剩 =Q 总 -Q 加速 -Q 匀速 -Q 制动 -Q 静止 , Remaining flight time of the drone Remaining driving distance Compare the Manhattan distance L between the drone and the controller with the remaining driving distance. If S 剩 -L≤0.2S 剩 , then a return power warning is issued. If S 剩 ≤L, then look for a safe place to land.

2. According to claim 1, a visual synchronization UAV flight control method is characterized in that: Collect the analog signal of the drone sensor, and obtain the drone digital signal through analog-to-digital conversion, including: Combined video sensors, with overlapping sensor viewing angles, can obtain multiple sets of drone visual signals; Locate the drone through the position sensor and obtain the drone's spatial position signal; Detect the real-time power of the drone through power detection and obtain the remaining power signal of the drone; The wind direction, wind force and wind speed of the current drone are measured through the wind sensor to obtain the drone's wind signal; The signal acquired by the drone is converted into digital form through an analog-to-digital converter to obtain the digital signal of the drone.

3. The visual synchronization UAV flight control method according to claim 1 is characterized in that: The control end sends the drone control command, the control command performs neural network training, and deep learning obtains the drone control neural network, including: When affected by wind, the wind factor neural network is preset for training, and deep learning of wind utilization is performed to obtain a wind optimization control neural network; When the battery level is affected, the preset neural network training is performed on the remaining battery level of the drone, and deep learning is used to obtain the remaining battery level return control neural network; When the drone is affected by close contact, neural network training is performed on the close-range braking of the drone, and deep learning is used to obtain the contact distance braking control neural network; When affected by the signal transmission threshold, neural network training is performed on the edge control distance braking of the drone, and deep learning is used to obtain the control distance braking control neural network.

4. A visually synchronized UAV flight control system, characterized in that: The system comprises: The data acquisition module is used to collect the analog signals of the drone sensors and convert the analog signals into digital signals to obtain the digital signals of the drone; The transmission processing module is used to transmit the UAV digital signal to the control terminal and obtain the UAV visualization data after processing; Data warning module, used to set the threshold of drone data collection, monitor and control the threshold, and obtain warning information when the threshold is exceeded; The learning control module is used to receive the drone control instructions sent by the control end, perform neural network training on the control instructions, and obtain the drone control neural network through deep learning; The transmission processing module comprises: The transmission and distribution module is used to transmit the UAV digital signal to the control end, package and distribute the digital signal at the control end, and send the digital signal to be processed to the corresponding processor; The position processing module is used to extract the spatial position signal of the UAV, establish a spatial coordinate system with the position of the control end as the coordinate origin, and obtain the UAV digital processing coordinate system; The video processing module is used to combine the video sensor's viewing angle to locate the spatial coordinates, intercept the spatial coordinates of the intersection, fuse the intersection data, and obtain the UAV wide-angle video digital signal; The distance processing module is used to combine the drone digital processing coordinate system with the drone wide-angle video digital signal, and obtain the digital signal of the distance between the drone and the target point through spatial coordinate calculation; the drone takes off from the control end, and obtains the length, width and height data of the current distance of the drone from the origin of the control end through the position sensor. Assuming the spatial coordinates of the drone are (x, y, z), the Manhattan distance of the drone flying back to the control end is L, L = |x| + |y| + |z|, The straight-line distance between the drone and the origin of the control end is r. Put the detection target into the spatial coordinate system, the center point of the monitoring target is (x1, y1, z1), the distance between the drone and the target is L'=|x-x1|+|y-y1|+|z-z1|, when L'=0, the drone contacts and collides with the target point; The visual conversion module is used to convert the processed drone digital signal into digital-to-analog through a digital-to-analog converter, and the data is displayed on the LCD display of the control end to obtain the visual image of the drone's wide-angle synchronous video, real-time power, wind speed and direction, and the distance between the drone and the target point; The data early warning step of the data early warning module includes: Assume the weight of the drone is m, the vertical torque of the drone is F1, the working power corresponding to torque F1 is P1, the horizontal longitudinal torque is F2, the working power corresponding to torque F2 is P2, the horizontal lateral torque is F3, the working power corresponding to torque F2 is P3, the real-time wind force is F4, and the wind speed v 风 The angle between the wind force F4 and the vertical upward direction of the UAV is w1, the angle between F4 and the horizontal longitudinal outward direction of the UAV is w2, and the angle between F4 and the horizontal lateral outward direction of the UAV is w3. The UAV operation time is t. The vertical force F of the wind force F4 dispersed to the UAV is 风垂 =F4*cosw1, the wind force F4 dispersed to the horizontal longitudinal force F of the drone 风纵 =F4*cosw2, F3 is the force F distributed to the horizontal longitudinal direction of the drone 风横 =F4*cosw3, In a windy environment, the vertical force F on the drone 垂 =F 风垂 +mg, g is the acceleration due to gravity, when F1=F 垂 When F1≠F 垂 When the UAV moves in the vertical direction with an acceleration a 垂 To speed up, In a windy environment, the horizontal longitudinal force F of the drone is 纵 =F 风纵 +F2, when F 纵 = 0, the drone maintains horizontal longitudinal braking. 纵 ≠0, the UAV moves in the horizontal longitudinal direction with an acceleration a 纵 To speed up, In a windy environment, the horizontal lateral force F of the drone is 横 =F 风横 +F3, when F 横 = 0, the drone maintains horizontal lateral braking. 横 ≠0, the UAV moves horizontally with an acceleration a 横 To speed up, The time for the drone to accelerate is t 加速 , uniform time is t 匀 , braking time is t 制 , the target point coordinates are (x1, y1, z1), the acceleration time distance, the speed of the drone before acceleration is v0, and the speed after acceleration is v 末 , Power consumption during acceleration Power consumption during constant speed The emergency braking time of the drone is t 制 After braking, the aircraft is stationary in the air, and the braking energy Vertical braking distance Horizontal longitudinal braking distance Horizontal braking distance When the drone is stationary, the vertical torque is the opposite force of the aircraft's gravity and the vertical wind force, the horizontal longitudinal torque is the opposite force of the wind's horizontal longitudinal force, and the horizontal lateral torque is the opposite force of the wind's horizontal lateral force. First, set the distance threshold between the drone and the control terminal, and give a prompt when the distance exceeds the threshold to obtain control distance warning information; The straight-line distance between the drone and the origin of the control end is r. When r≥200 meters, the control is too far and a warning message is issued; Then, the contact distance threshold between the drone and the target is set, and the distance between the drone and the target is monitored in real time. When the contact distance threshold is reached, the contact distance warning information is obtained; the Manhattan distance of the drone target in each direction is compared with the braking distance. When the distance difference in the three directions is close to 0, that is, |x-x1|-S 垂制 →0, and |y-y1|-S 纵制 →0, and |z-z1|-S 横制 →0, conduct contact alarm warning; After that, set the wind speed threshold of the drone. When the actual wind speed exceeds the rotor torque range of the drone, that is, when F1-F 垂 ≤0.3F1, or F2-F 纵 ≤0.3F2, or F3-F 横 ≤0.3F3, a wind danger control warning is issued; when F1≤F 垂 , or F2≤F 纵 , or F3≤F 横 When the wind is out of control, a flight ban warning will be issued; Finally, set the drone power threshold to obtain the current drone return power required. If the drone return power exceeds the return power, a return power warning will be issued, and the remaining power of the drone is Q 剩 =Q 总 -Q 加速 -Q 匀速 -Q 制动 -Q 静止 , Remaining flight time of the drone Remaining driving distance Compare the Manhattan distance L between the drone and the controller with the remaining driving distance. If S 剩 -L≤0.2S 剩 , then a return power warning is issued. If S 剩 ≤L, then look for a safe place to land.

5. A visually synchronized UAV flight control system according to claim 4, characterized in that: The data acquisition module comprises: Video data module, used to combine video sensors, collect video data and obtain multiple sets of drone visual signals; Position data module, used to locate the drone and obtain the drone's spatial position signal; The power detection module is used to detect the real-time power of the drone and obtain the remaining power signal of the drone; The wind force detection module is used to measure the wind direction, wind force and wind speed of the current drone and obtain the drone wind force signal; The signal conversion module performs analog-to-digital conversion on the signal acquired by the drone to obtain the drone digital signal.

6. A visually synchronized UAV flight control system according to claim 4, characterized in that: The learning control module comprises: The wind power utilization control module is used to preset wind power factors for neural network training, conduct deep learning on wind power utilization, and obtain a wind power optimization control neural network; The battery return control module is used to perform preset neural network training on the remaining battery of the drone, and obtain the remaining battery return control neural network through deep learning; The contact distance control module is used to train the neural network for close-range braking of the drone and obtain the contact distance braking control neural network through deep learning; The edge distance control module is used to train the neural network for the edge control distance braking of the drone, and to obtain the control distance braking control neural network through deep learning.

Images