Unmanned aerial vehicle with detection and throwing functions

Abstract

The invention discloses an unmanned aerial vehicle with detection and throwing functions. The unmanned aerial vehicle comprises a flight control module, a power module, an image acquisition module, a digital image transmission communication module, a positioning navigation system and a throwing device, the designed unmanned aerial vehicle system takes off manually, the ground station uploads tasks, task modes are switched after the unmanned aerial vehicle system arrives at a task area, and the throwing device driven by the steering engine is used for completing automatic throwing tasks. The unmanned aerial vehicle is stable in function, the master control and the sensor work normally, and autonomous detection and launching tasks can be completed according to requirements. The unmanned aerial vehicle designed by the invention shows very high detection accuracy and instant launching accuracy, and realizes the functions of autonomous flight, route planning, image acquisition, route planning, image processing, target recognition, accurate launching and the like of the unmanned aerial vehicle. The unmanned aerial vehicle designed by the invention has the performance advantages of large coverage range, long air control time, high timeliness and the like.

Description

Technical Field

[0001] The present invention relates to the field of unmanned aerial vehicles (UAVs), and in particular to a UAV with both detection and throwing functions. Background Art

[0002] UAV, or Unmanned Aerial Vehicle, is an unmanned aircraft that is self-powered, can carry various electronic equipment, perform multiple tasks, and can be controlled by an operator through a radio remote control device or by an automatic remote control program device that comes with the UAV. The biggest feature of a UAV is that it does not require a pilot to drive and operate it in the fuselage, while being able to perform almost all the functions of a manned aircraft. Therefore, a UAV does not need to be equipped with a pilot seat, a joystick, or other devices that enable the pilot to complete the aircraft's control functions, nor does it need to be equipped with a cockpit pressurization system or other devices that are necessary for a manned aircraft to maintain the pilot's life. UAVs are characterized by light weight, simple structure, easy operation, low cost of use, and safe and efficient use.

[0003] For nearly a century, drones have played an important role in the two major fields of civil and scientific research with their advantages of low loss, zero casualties and high maneuverability. However, existing drones still have defects such as inaccurate delivery, short air control time and small flight coverage area. Summary of the invention

[0004] The purpose of the present invention is to provide a drone with both detection and throwing functions, comprising: a flight control module, a power module, an image acquisition module, a digital image transmission communication module, a positioning and navigation system, and a throwing device.

[0005] After receiving the flight command, the flight control module generates a signal for controlling the flight trajectory and posture of the drone and an object delivery signal, transmits the signal for controlling the flight trajectory and posture of the drone to the power module, and transmits the object delivery signal to the throwing device.

[0006] The signals for controlling the flight trajectory and posture of the UAV include a steering control signal, a steering angle control signal, and a flight speed control signal.

[0007] When the power module receives the steering control signal and the steering angle control signal, it drives the drone to turn and adjusts the steering angle of the drone according to the steering angle control signal.

[0008] When the power module receives the flight speed control signal, it adjusts the flight speed of the drone.

[0009] The image acquisition module is used to acquire images of the environment where the drone is located and images below the drone, and transmit the acquired images to the digital image transmission communication module.

[0010] The digital image transmission communication module is used to transmit the image information collected by the image acquisition module back to the ground station.

[0011] The ground station processes the received image information, generates flight instructions, and transmits the flight instructions to the flight control module of the UAV.

[0012] The positioning and navigation system is used to locate the object throwing point.

[0013] After receiving the object throwing signal, the throwing device throws the object carried by the drone.

[0014] The drone is equipped with a power module, an image acquisition module, a flight control module, a digital image transmission communication module, a positioning and navigation system, and a throwing device.

[0015] Furthermore, the drone is also equipped with a power module.

[0016] The power supply module supplies power to the flight control module, the power module, the image acquisition module, the digital image transmission communication module, the positioning and navigation system, and the throwing device.

[0017] Furthermore, the image information collected by the image acquisition module is transmitted back by the digital image transmission communication module using an opencv digital recognition algorithm.

[0018] Further, the steps of the opencv digital recognition algorithm include:

[0019] 1) Convert the RGB information of the acquired image into the HSV color space to identify the color of the target.

[0020] 2) Extract the minimum circumscribed rectangle of the red and blue color areas in the HSV color space and convert it into a binary image to obtain a rectangular binary image.

[0021] 3) Store the rectangular binary image in the defined vector container.

[0022] 4) Prepare template images of ten numbers from 0 to 9 and convert them into binary images to obtain a template binary image.

[0023] 5) Compare the rectangular binary image in the vector container with the template binary image pixel by pixel to obtain the similarity.

[0024] 6) Compare the similarity between each rectangular binary image and the template binary image, and the template image number corresponding to the highest similarity value is the recognized number.

[0025] Furthermore, the default coordinate system of the drone is set to be consistent with the map coordinate system.

[0026] The positioning and navigation system adopts a moving point comparison correction algorithm to perform deviation correction.

[0027] Furthermore, the steps of the moving point comparison correction algorithm include:

[0028] 1) Set the take-off position of the drone to (x0, y0) and the target point position to (x i ,y i ), i is the target point ordinal number, i = 1, 2, ..., n, n is the total number of target points.

[0029] 2) The drone flies to the target point (x i ,y i ), the actual position reached by the drone is (x′ i , y′ i ).

[0030] 3) The recognition point of the drone for the target image is (x″ i , y″ i ).

[0031] 4) Compare the actual position (x′ i , y′ i ) and drone identification point (x″ i , y″ i ) to correct the position deviation of the UAV.

[0032] The technical effect of the present invention is unquestionable. The UAV system designed by the present invention takes off manually, the ground station uploads the task, and switches the task mode after arriving at the task area, realizing functions such as autonomous flight, ground reconnaissance, data return, image processing, target recognition, and autonomous delivery. The automatic delivery task is completed by using the throwing device driven by the steering gear. The UAV of the present invention has stable functions, the main control and the sensor work normally, and can complete the tasks of autonomous detection and delivery as required.

[0033] The drone designed by the present invention exhibits high detection accuracy and instant delivery accuracy, and realizes the functions of autonomous flight, route planning, image acquisition, route planning, image processing, target recognition, and precise delivery of the drone. The drone designed by the present invention has the performance advantages of large coverage, long air control time, and strong timeliness. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 A schematic diagram of a drone with both detection and throwing functions according to the present invention;

[0035] Figure 2 A top view of the drone with both detection and throwing functions of the present invention;

[0036] Figure 3 is an overall block diagram of the hardware system of the present invention;

[0037] Figure 4It is the overall working flow chart of the present invention. DETAILED DESCRIPTION

[0038] The present invention is further described below in conjunction with the embodiments, but it should not be understood that the above subject matter of the present invention is limited to the following embodiments. Without departing from the above technical ideas of the present invention, various substitutions and changes are made according to the common technical knowledge and customary means in the art, which should all be included in the protection scope of the present invention.

[0039] Embodiment 1:

[0040] See also Figures 1 to 4 , a UAV with both detection and throwing functions, including: a flight control module, a power module, an image acquisition module, a digital image transmission communication module, a positioning and navigation system and a throwing device.

[0041] After receiving the flight command, the flight control module generates a signal to control the flight trajectory and attitude of the drone and a signal to drop items, and transmits the signal to control the flight trajectory and attitude of the drone to the power module, and transmits the signal to drop items to the throwing device. The flight control module uses the Leixun V5+ flight controller, which is based on the Pixhawk FMUv5 design standard, with mature technology, simple debugging, flexibility and reliability, and good flight effect.

[0042] The signals for controlling the flight trajectory and posture of the UAV include a steering control signal, a steering angle control signal, and a flight speed control signal.

[0043] When the power module receives the steering control signal and the steering angle control signal, it drives the drone to turn and adjusts the steering angle of the drone according to the steering angle control signal.

[0044] When the power module receives the flight speed control signal, it controls the flight speed of the drone. The power module includes SUNNYSKY V2814 KV800 four-axis brushless motor, ATG 1045 forward and reverse propellers, and Haoying Letian X-Rotor40A electronic speed controller, and the overall power is sufficient.

[0045] The image acquisition module is used to collect images of the environment where the drone is located and the image below the drone, and transmit the collected images to the digital image transmission communication module. The image acquisition module uses the OV OS12D40 color image sensor, which can provide 11.3 million pixel (4512x2512) image signals at a maximum speed of 60fps, and has an EIS anti-shake function for acquiring ground images.

[0046] The digital image transmission communication module is used to transmit the image information collected by the image acquisition module back to the ground station. The digital image transmission communication module is a Homer digital image transmission integrated module, which is used to transmit the image information collected by the OV OS12D40 color image sensor.

[0047] The ground station processes the received image information, generates flight instructions, and transmits the flight instructions to the flight control module of the UAV.

[0048] The positioning and navigation system is used to locate the object throwing point. The positioning and navigation system adopts the NEO V2 GNSS positioning module, which has strong anti-interference ability, fast satellite search speed, can receive three global navigation satellite systems at the same time, and has a tracking and navigation accuracy of 167dbm.

[0049] After receiving the signal for dropping items, the throwing device throws the items carried by the drone. The throwing device is designed independently. The long sides of two wooden boards are glued to the bottom of the fuselage. The middle of the two wooden boards is supported by an iron rod, which is fixed to the steering arm. When the steering arm does not rotate, the two wooden boards are closed; when the steering arm rotates 90°, the two wooden boards are opened. To ensure that the water bottle can fall when the wooden board is opened, an elastic belt is used to provide downward elastic force when loading the water bottle.

[0050] The drone is equipped with a power module, an image acquisition module, a flight control module, a digital image transmission communication module, a positioning and navigation system, and a throwing device.

[0051] We use the MFD Mini Halo 1600 as the platform for the drone system. This model has a high lift-to-drag ratio and good medium and low speed performance, which can well meet the mission requirements. The camera, flight control module, and digital image transmission system are installed inside the fuselage, and a throwing device is installed near the bottom. To ensure the structural strength, wooden boards are added for reinforcement.

[0052] The OS12D40 three-axis anti-shake gimbal camera module is equipped with the iCatch V39M chip, which supports H.264-encoded videos of up to 4K@60FPS, 4K@30FPS, and 1080P@120FPS. It uses the OV OS12D40 color image sensor, which can provide 11.3 million pixel (4512x2512) image signals at a maximum speed of 60fps, and has EIS anti-shake and multi-axis mechanical anti-shake functions.

[0053] V39M motherboard + wide-angle distortion-free module, equipped with iCatch V39M chip, supports up to 4K@60FPS, 4K@30FPS, 1080P@120FPS H.264 encoded video, uses SONY IMX377 high-quality CMOS sensor, with 12 million effective pixels and 100-degree zero-distortion wide-angle lens. Supports multi-axis EIS image stabilization function.

[0054] Solution selection: The image acquisition module of this mission has a working height of 15-20m, and the image sensor resolution is required to be high. Considering the windy weather at the site, the OS12D40 three-axis anti-shake gimbal module supports multi-axis mechanical image stabilization in addition to the EIS image stabilization function. Choosing the OS12D40 three-axis anti-shake gimbal module can provide higher quality image information and increase system reliability. The disadvantage is that the structure is relatively heavy.

[0055] The Homer image data transmission system has the advantages of small size, low power consumption, and high sensitivity. In actual use, the image delay can be controlled within 400-500ms.

[0056] In summary, the combination of the OS12D40 three-axis anti-shake gimbal module and the Homer image data transmission system can meet the needs of this mission to collect high-resolution, low-latency images.

[0057] Solution 1: Homer digital and image transmission module

[0058] The Homer image data transmission system operates in the 5.1GHz to 5.9GHz frequency band, using wireless communication to transmit network information and serial port transparent information. With one AP / STA node, the speed between nodes can reach up to 1600Mbps; it is very suitable for large-scale data transmission and exchange.

[0059] Solution 2: GSLINK-A2 integrated digital and image transmission module

[0060] GSLINK-A2 integrates MAVLINK data and HDMI video through 4G network for unlimited distance transmission, and displays flight data and high-definition images in real time on MissionPlaner ground station or GQC ground station.

[0061] GSLINK uses high-definition H264 encoding technology and supports up to 1080P25fps video input. Ideally, data communication delay is kept within 100ms and video delay is around 250ms.

[0062] Solution selection: Solution 1 requires connecting the air end to the drone, while Solution 2 requires higher stability of the 4G signal. Solution 1's air end can be connected to the OS12D40 three-axis anti-shake gimbal module, and there is no need to configure a voltage stabilization module separately. The disadvantage is that the structure weight is increased; the 4G signal at the site is unstable and intermittent. After comprehensive consideration, Solution 1 is selected.

[0063] The drone is also equipped with a power module. The power module uses a Huapai infinity 6S battery with a voltage of 22.2V, a capacity of 5200mAh, and a discharge rate of 55C to provide power for the motor and other modules. The design also uses a 5A output DC voltage regulator and step-down module, through which the power supply is realized to the flight control module, the OS12D40 three-axis anti-shake gimbal module, and the Homer digital transmission and image transmission integrated module to ensure the stable and coordinated operation of each unit of the drone.

[0064] The power supply module supplies power to the flight control module, the power module, the image acquisition module, the digital image transmission communication module, the positioning and navigation system, and the throwing device.

[0065] The image information collected by the image acquisition module is transmitted back by the digital image transmission communication module using the opencv digital recognition algorithm.

[0066] The steps of the opencv digital recognition algorithm include:

[0067] 1) Convert the RGB information of the acquired image into the HSV color space to identify the color of the target.

[0068] 2) Extract the minimum circumscribed rectangle of the red and blue color areas in the HSV color space and convert it into a binary image to obtain a rectangular binary image.

[0069] 3) Store the rectangular binary image in the defined vector container.

[0070] 4) Prepare template images of ten numbers from 0 to 9 and convert them into binary images to obtain a template binary image.

[0071] 5) Compare the rectangular binary image in the vector container with the template binary image pixel by pixel to obtain the similarity.

[0072] 6) Compare the similarity between each rectangular binary image and the template binary image, and the template image number corresponding to the highest similarity value is the recognized number.

[0073] Initially, we compared the actual position of the drone with the predetermined position to make deviation corrections. However, the drone believed that its current position was the target point we gave it, that is, the drone believed that its position had almost no deviation, and the correction amount was very random and could not work stably. Moreover, when making deviation corrections, the correction amount can only be accurately corrected in the same coordinate system. At the beginning, the default coordinate system on the drone did not correspond to the map coordinate system, which also made it impossible to make stable corrections during flight.

[0074] Set the drone's default coordinate system to be consistent with the map coordinate system.

[0075] The positioning and navigation system adopts a moving point comparison correction algorithm to perform deviation correction.

[0076] The steps of the moving point comparison correction algorithm include:

[0077] 1) Set the take-off position of the drone to (x0, y0) and the target point position to (x i ,y i ), i is the target point ordinal number, i = 1, 2, ..., n, n is the total number of target points.

[0078] 2) The drone flies to the target point (x i ,y i ), the actual position reached by the drone is (x′ i , y′ i ).

[0079] 3) The recognition point of the drone for the target image is (x″ i , y″ i ).

[0080] 4) Compare the actual position (x′ i , y′ i ) and drone identification point (x″ i , y″ i ) to correct the position deviation of the UAV.

[0081] Embodiment 2:

[0082] See also Figures 1 to 4 , a UAV with both detection and throwing functions, including: a flight control module, a power module, an image acquisition module, a digital image transmission communication module, a positioning and navigation system and a throwing device.

[0083] After receiving the flight command, the flight control module generates a signal to control the flight trajectory and attitude of the drone and a signal to drop items, and transmits the signal to control the flight trajectory and attitude of the drone to the power module, and transmits the signal to drop items to the throwing device. The flight control module uses the Leixun V5+ flight controller, which is based on the Pixhawk FMUv5 design standard, with mature technology, simple debugging, flexibility and reliability, and good flight effect.

[0084] The signals for controlling the flight trajectory and posture of the UAV include a steering control signal, a steering angle control signal, and a flight speed control signal.

[0085] When the power module receives the steering control signal and the steering angle control signal, it drives the drone to turn and adjusts the steering angle of the drone according to the steering angle control signal.

[0086] When the power module receives the flight speed control signal, it controls the flight speed of the drone. The power module includes SUNNYSKY V2814 KV800 four-axis brushless motor, ATG 1045 forward and reverse propellers, and Haoying Letian X-Rotor40A electronic speed controller, and the overall power is sufficient.

[0087] The image acquisition module is used to collect images of the environment where the drone is located and the image below the drone, and transmit the collected images to the digital image transmission communication module. The image acquisition module uses the OV OS12D40 color image sensor, which can provide 11.3 million pixel (4512x2512) image signals at a maximum speed of 60fps, and has an EIS anti-shake function for acquiring ground images.

[0088] The digital image transmission communication module is used to transmit the image information collected by the image acquisition module back to the ground station. The digital image transmission communication module is a Homer digital image transmission integrated module, which is used to transmit the image information collected by the OV OS12D40 color image sensor.

[0089] The ground station processes the received image information, generates flight instructions, and transmits the flight instructions to the flight control module of the UAV.

[0090] The positioning and navigation system is used to locate the object throwing point. The positioning and navigation system adopts the NEO V2 GNSS positioning module, which has strong anti-interference ability, fast satellite search speed, can receive three global navigation satellite systems at the same time, and has a tracking and navigation accuracy of 167dbm.

[0091] After receiving the signal for dropping items, the throwing device throws the items carried by the drone. The throwing device is designed independently. The long sides of two wooden boards are glued to the bottom of the fuselage. The middle of the two wooden boards is supported by an iron rod, which is fixed to the steering arm. When the steering arm does not rotate, the two wooden boards are closed; when the steering arm rotates 90°, the two wooden boards are opened. To ensure that the water bottle can fall when the wooden board is opened, an elastic belt is used to provide downward elastic force when loading the water bottle.

[0092] The drone is equipped with a power module, an image acquisition module, a flight control module, a digital image transmission communication module, a positioning and navigation system, and a throwing device.

[0093] We use the MFD Mini Halo 1600 as the platform for the drone system. This model has a high lift-to-drag ratio and good medium and low speed performance, which can well meet the mission requirements. The camera, flight control module, and digital image transmission system are installed inside the fuselage, and a throwing device is installed near the bottom. To ensure the structural strength, wooden boards are added for reinforcement.

[0094] The OS12D40 three-axis anti-shake gimbal camera module is equipped with the iCatch V39M chip, which supports H.264-encoded videos of up to 4K@60FPS, 4K@30FPS, and 1080P@120FPS. It uses the OV OS12D40 color image sensor, which can provide 11.3 million pixel (4512x2512) image signals at a maximum speed of 60fps, and has EIS anti-shake and multi-axis mechanical anti-shake functions.

[0095] V39M motherboard + wide-angle distortion-free module, equipped with iCatch V39M chip, supports up to 4K@60FPS, 4K@30FPS, 1080P@120FPS H.264 encoded video, uses SONY IMX377 high-quality CMOS sensor, with 12 million effective pixels and 100-degree zero-distortion wide-angle lens. Supports multi-axis EIS image stabilization function.

[0096] Solution selection: The image acquisition module of this mission has a working height of 15-20m, and the image sensor resolution is required to be high. Considering the windy weather at the site, the OS12D40 three-axis anti-shake gimbal module supports multi-axis mechanical image stabilization in addition to the EIS image stabilization function. Choosing the OS12D40 three-axis anti-shake gimbal module can provide higher quality image information and increase system reliability. The disadvantage is that the structure is relatively heavy.

[0097] The Homer image data transmission system has the advantages of small size, low power consumption, and high sensitivity. In actual use, the image delay can be controlled within 400-500ms.

[0098] In summary, the combination of the OS12D40 three-axis anti-shake gimbal module and the Homer image data transmission system can meet the needs of this mission to collect high-resolution, low-latency images.

[0099] Solution 1: Homer digital and image transmission module

[0100] The Homer image data transmission system operates in the 5.1GHz to 5.9GHz frequency band, using wireless communication to transmit network information and serial port transparent information. With one AP / STA node, the speed between nodes can reach up to 1600Mbps; it is very suitable for large-scale data transmission and exchange.

[0101] Solution 2: GSLINK-A2 integrated digital and image transmission module

[0102] GSLINK-A2 integrates MAVLINK data and HDMI video through 4G network for unlimited distance transmission, and displays flight data and high-definition images in real time on MissionPlaner ground station or GQC ground station.

[0103] GSLINK uses high-definition H264 encoding technology and supports up to 1080P25fps video input. Ideally, data communication delay is kept within 100ms and video delay is around 250ms.

[0104] Solution selection: Solution 1 requires connecting the air end to the drone, while Solution 2 requires higher stability of the 4G signal. Solution 1's air end can be connected to the OS12D40 three-axis anti-shake gimbal module, and there is no need to configure a voltage stabilization module separately. The disadvantage is that the structure weight is increased; the 4G signal at the site is unstable and intermittent. After comprehensive consideration, Solution 1 is selected.

[0105] Embodiment 3:

[0106] A drone with both detection and throwing functions, the main structure of which is shown in Example 2, wherein the drone is also equipped with a power module. The power module uses a Huapai infinity 6S battery with a voltage of 22.2V, a capacity of 5200mAh, and a discharge rate of 55C to provide power for the motor and other modules. The design also uses a 5A output DC voltage regulator and step-down module, through which the power supply is realized to the flight control module, the OS12D40 three-axis anti-shake gimbal module, and the Homer digital transmission and image transmission integrated module to ensure the stable and coordinated operation of each unit of the drone.

[0107] The power supply module supplies power to the flight control module, the power module, the image acquisition module, the digital image transmission communication module, the positioning and navigation system, and the throwing device.

[0108] Embodiment 4:

[0109] A drone with both detection and throwing functions, the main structure of which is shown in Example 2, wherein the image information collected by the image acquisition module transmitted back by the digital image transmission communication module adopts the opencv digital recognition algorithm.

[0110] Embodiment 5:

[0111] A drone with both detection and throwing functions, the main structure of which is shown in Example 4, wherein the steps of the opencv digital recognition algorithm include:

[0112] 1) Convert the RGB information of the acquired image into the HSV color space to identify the color of the target.

[0113] 2) Extract the minimum circumscribed rectangle of the red and blue color areas in the HSV color space and convert it into a binary image to obtain a rectangular binary image.

[0114] 3) Store the rectangular binary image in the defined vector container.

[0115] 4) Prepare template images of ten numbers from 0 to 9 and convert them into binary images to obtain a template binary image.

[0116] 5) Compare the rectangular binary image in the vector container with the template binary image pixel by pixel to obtain the similarity.

[0117] 6) Compare the similarity between each rectangular binary image and the template binary image, and the template image number corresponding to the highest similarity value is the recognized number.

[0118] Embodiment 6:

[0119] A drone with both detection and throwing functions, the main structure is shown in Example 2, wherein the actual position of the drone is compared with the predetermined position to make deviation correction at first, but for the drone, it thinks that its current position is the target point given by us, that is, the drone thinks that its position has almost no deviation, and its correction amount is very random and cannot work stably. And when making deviation correction, the correction amount can only be accurately corrected in the same coordinate system, and the default coordinate system on the drone does not correspond to the map coordinate system at the beginning, which also makes it impossible to make stable corrections during flight.

[0120] Set the drone's default coordinate system to be consistent with the map coordinate system.

[0121] The positioning and navigation system adopts a moving point comparison correction algorithm to perform deviation correction.

[0122] Embodiment 7:

[0123] A drone with both detection and throwing functions, the main structure of which is shown in Example 6, wherein the steps of the moving point comparison correction algorithm include:

[0124] 1) Set the take-off position of the drone to (x0, y0) and the target point position to (x i ,y i ), i is the target point ordinal number, i = 1, 2, ..., n, n is the total number of target points.

[0125] 2) The drone flies to the target point (x i ,y i ), the actual position reached by the drone is (x′ i , y′ i ).

[0126] 3) The recognition point of the drone for the target image is (x″ i , y″ i ).

[0127] 4) Compare the actual position (x′ i , y′ i ) and drone identification point (x″ i, y″ i ) to correct the position deviation of the UAV.

[0128] Embodiment 8:

[0129] A drone with both detection and throwing functions, the content is as follows:

[0130] The present invention plans to use a fixed-wing UAV with a conventional layout, good low-speed performance, and dimensions that meet the requirements. In combination with the requirements of the topic, the team plans to install image sensors and image data transmission systems on the UAV to realize data collection and transmission of ground targets, and complete autonomous flight based on QGroundcontrol to plan routes, thus realizing autonomous flight, image collection, target recognition, and autonomous delivery of the UAV. The main work done is as follows: First, the mechanical structure and power system of the UAV are selected and designed according to the requirements of the competition mission; the image acquisition module, data return module, and throwing device are selected; based on OpenCV, the software of the UAV system is designed and simulated. Finally, the assembly and mission verification of the UAV are completed on the basis of the above. The test results show that the UAV exhibits high reconnaissance accuracy and instant delivery accuracy, and realizes the functions of autonomous flight, route planning, image acquisition, route planning, image processing, target recognition, and precise delivery of the UAV.

[0131] The following is a demonstration and selection of the solutions involved in the system.

[0132] Selection of image acquisition module

[0133] Solution 1: OS12D40 three-axis anti-shake gimbal module

[0134] The OS12D40 three-axis anti-shake gimbal camera module is equipped with the iCatch V39M chip, which supports H.264-encoded videos of up to 4K@60FPS, 4K@30FPS, and 1080P@120FPS. It uses the OV OS12D40 color image sensor, which can provide 11.3 million pixel (4512x2512) image signals at a maximum speed of 60fps, and has EIS anti-shake and multi-axis mechanical anti-shake functions.

[0135] Solution 2: V39M motherboard + wide-angle distortion-free module

[0136] Equipped with iCatch V39M chip, it supports up to 4K@60FPS, 4K@30FPS, 1080P@120FPS H.264 encoded videos, uses SONY IMX377 high-quality CMOS sensor, has 12 million effective pixels, and 100-degree zero-distortion wide-angle lens. It supports multi-axis EIS image stabilization function.

[0137] Solution selection: The image acquisition module of this mission has a working height of 15-20m, and the image sensor resolution is required to be high. Considering the windy weather at the site, the OS12D40 three-axis anti-shake gimbal module supports multi-axis mechanical image stabilization in addition to the EIS image stabilization function. Choosing the OS12D40 three-axis anti-shake gimbal module can provide higher quality image information and increase system reliability. The disadvantage is that the structure is relatively heavy.

[0138] Selection of data return module

[0139] The Homer image data transmission system has the advantages of small size, low power consumption, and high sensitivity. In actual use, the image delay can be controlled within 400-500ms.

[0140] In summary, the combination of the OS12D40 three-axis anti-shake gimbal module and the Homer image data transmission system can meet the needs of this mission to collect high-resolution, low-latency images.

[0141] Solution 1: Homer digital and image transmission module

[0142] The Homer image data transmission system operates in the 5.1GHz to 5.9GHz frequency band, using wireless communication to transmit network information and serial port transparent information. With one AP / STA node, the speed between nodes can reach up to 1600Mbps; it is very suitable for large-scale data transmission and exchange.

[0143] Solution 2: GSLINK-A2 integrated digital and image transmission module

[0144] GSLINK-A2 integrates MAVLINK data and HDMI video through 4G network for unlimited distance transmission, and displays flight data and high-definition images in real time on MissionPlaner ground station or GQC ground station.

[0145] GSLINK uses high-definition H264 encoding technology and supports up to 1080P25fps video input. Ideally, data communication delay is kept within 100ms and video delay is around 250ms.

[0146] Solution selection: Solution 1 requires connecting the air end to the drone, while Solution 2 requires higher stability of the 4G signal. Solution 1's air end can be connected to the OS12D40 three-axis anti-shake gimbal module, and there is no need to configure a voltage stabilization module separately. The disadvantage is that the structure weight is increased; the 4G signal at the site is unstable and intermittent. After comprehensive consideration, Solution 1 is selected.

[0147] Design and calculation

[0148] openCV digital recognition algorithm

[0149] First, the image sent back by the drone is used to identify the color of the target using the RGB and HSV conversion algorithm based on OpenCV. The RGB information of the target collected by the camera is converted into the HSV color space to identify the color of the target. The minimum enclosing rectangle of the red and blue color areas is extracted and converted into a binary image. Finally, the obtained rectangles are stored in the defined vector container.

[0150] Prepare template images of ten numbers from 0 to 9 in advance, convert them into binary images, and compare the segmented image with the template image by traversing pixels to obtain the similarity. The number in the template image with the highest similarity is the recognized number.

[0151] Moving point comparison correction algorithm

[0152] Initially, we compared the actual position of the drone with the predetermined position to make deviation corrections. However, the drone believed that its current position was the target point we gave it, that is, the drone believed that its position had almost no deviation, and the correction amount was very random and could not work stably. Moreover, when making deviation corrections, the correction amount can only be accurately corrected in the same coordinate system. At the beginning, the default coordinate system on the drone did not correspond to the map coordinate system, which also made it impossible to make stable corrections during flight.

[0153] After improvement, the default coordinate system of the UAV is modified in the control algorithm to keep it consistent with the map coordinate system. The UAV takes off at (x0, y0) and flies to the first target point (x1, y1). When the UAV flies to (x1, y1), there will be a certain deviation. At this time, the actual position of the UAV is (x′1, y′1), and its target image recognition point is (x″1, y″1). These two points are dynamic points. These two dynamic points are compared and the deviation is corrected. While correcting the deviation, the altitude is lowered and the pod is dropped. After the pod is folded, it flies from the current position to the next target point (x2, y2). The UAV position is updated to (x′2, y′2), and the image recognition point is (x″2, y″2), and then compared and corrected.

[0154] UAV hardware design

[0155] UAV mechanical structure design

[0156] The overall mechanical structure of the drone is as follows: Figure 1 and Figure 2 As shown:

[0157] We use the MFD Mini Halo 1600 as the platform for the drone system. This model has a high lift-to-drag ratio and good medium and low speed performance, which can well meet the mission requirements. The camera, flight control module, and digital image transmission system are installed inside the fuselage, and a throwing device is installed near the bottom. To ensure the structural strength, wooden boards are added for reinforcement.

[0158] UAV hardware system design

[0159] The overall block diagram of the hardware system is as follows: Figure 3 As shown:

[0160] The system mainly includes seven units: flight control module, power module, image acquisition module, digital image transmission and communication module, positioning and navigation system, throwing device, and power module.

[0161] The flight control module uses the Leixun V5+ flight controller, which is based on the Pixhawk FMUv5 design standard and has mature technology, simple debugging, flexibility and reliability, and good flight effects.

[0162] The power module includes SUNNYSKY V2814 KV800 four-axis brushless motor, ATG 1045 forward and reverse propellers, and Haoying Letian X-Rotor40A electronic speed controller. The overall power is sufficient to meet the requirements of completing automatic detection and delivery tasks.

[0163] The image acquisition module uses the OV OS12D40 color image sensor, which can provide 11.3 million pixel (4512x2512) image signals at a maximum speed of 60fps and has EIS anti-shake function for acquiring ground images.

[0164] The digital and image transmission communication module is a Homer integrated digital and image transmission module, which is used to transmit image information collected by the OV OS12D40 color image sensor.

[0165] The positioning and navigation system uses the NEO V2 GNSS positioning module, which has strong anti-interference ability, fast satellite search speed, can receive three global navigation satellite systems at the same time, and has a tracking and navigation accuracy of 167dbm.

[0166] The throwing device is designed independently. The long sides of two wooden boards are glued to the bottom of the fuselage. The two wooden boards are supported by iron rods in the middle. The iron rods are fixed to the servo arm. When the servo arm does not rotate, the two wooden boards are closed; when the servo arm rotates 90 degrees, the two wooden boards are opened. To ensure that the water bottle can fall when the wooden board is opened, an elastic belt is used to provide downward elastic force when loading the water bottle.

[0167] Power module: This design uses Huapai Infinity 6S battery with voltage of 22.2V, capacity of 5200mAh and discharge rate of 55C to provide power for motors and other modules. The design also uses a 5A output DC voltage regulator and step-down module, which can provide stable power supply to the flight control module, OS12D40 three-axis anti-shake gimbal module, and Homer digital transmission and image transmission module, ensuring stable and coordinated operation of various units of the drone.

[0168] The overall workflow of each requirement of the UAV system is as follows Figure 4 shown.

[0169] Embodiment 9:

[0170] A drone with both detection and throwing functions, the test plan includes:

[0171] (1) First, test the mechanical properties of the drone, check whether all parts are tightly installed, whether there are any loose screws, loose bonding, unstable center of gravity, etc.

[0172] (2) Check the circuits of the image acquisition module, flight control module, and image data transmission module to see if the circuit connections of each unit are correct and whether there are any wrong soldering or cold soldering at the welding parts;

[0173] (3) Turn on the power supply and check whether each unit works normally after power is turned on, whether there are any abnormal conditions in each circuit, and prevent damage to the circuit and electronic components;

[0174] (4) Program debugging: Test the program in blocks, first the common parts and then the independent parts. After all tests are correct, test according to the requirements.

[0175] (5) Overall test: After the sub-module tests are completed, an overall test will be conducted at the experimental site, including autonomous flight, ground reconnaissance, data feedback, image processing, target recognition, autonomous delivery, etc., to complete the test of automatic reconnaissance and strike missions under the premise of meeting the competition requirements.

[0176] Test conditions: Check the circuit connection of the drone several times before testing to ensure that there are no wrong soldering or cold soldering. Run the flight control program and image processing program to correct any defects.

[0177] Test instruments: laptop computer, stopwatch, digital multimeter.

[0178] Test Results

[0179] (1)Digit recognition test results

[0180] The OS12D40 three-axis anti-shake gimbal module is connected to the Homer digital image transmission system air end, and the Homer digital image transmission system ground end is connected to the laptop computer. The distance between the two is 200m; the distance between the camera and the digital board is 20m. The test results are shown in Table 1:

[0181] Table 1. Results of the digital recognition test

[0182] frequency 1 2 3 4 5 6 7 8 Identify numbers 57 94 38 80 76 16 34 41 Delay(ms) 394 412 436 497 362 461 501 384 Is it correct? yes yes yes yes yes yes yes yes

[0183] (2) Delivery test results

[0184] Manually control the drone system to take off, and the ground station plans the mission area to pass over the patio. When approaching the mission area, switch the mission mode, and calculate the delivery time and delivery accuracy. The test results are shown in Table 2:

[0185] Table 2. Results of the test

[0186]

[0187] (3) Full process test results

[0188] Manually control the drone system to take off, manually switch to the mission mode when reaching the predetermined height, automatically scout the location area during the drone's automatic flight, observe whether the ground station displays the feedback image, complete the mission and request to return, and stop the timing after manual landing. Record the mission time, delivery accuracy, and feedback numbers. The test results are shown in Table 3:

[0189] Table 3. Results of the whole process test

[0190]

[0191] Test results analysis

[0192] Based on the above test results, the following conclusions can be drawn:

[0193] The drone’s flight functions are normal;

[0194] The drone’s automatic flight function is normal;

[0195] The drone image recognition and processing functions are normal;

[0196] Drone drop functions properly.

[0197] To sum up, this system design meets the requirements.

Claims

1. A drone with both detection and throwing functions, characterized in that: include: Flight control module, power module, image acquisition module, digital image transmission and communication module, positioning and navigation system and throwing device; After receiving the flight command, the flight control module generates a signal for controlling the flight trajectory and posture of the drone and an object delivery signal, transmits the signal for controlling the flight trajectory and posture of the drone to the power module, and transmits the object delivery signal to the throwing device. The signals for controlling the flight trajectory and attitude of the UAV include a steering control signal, a steering angle control signal, and a flight speed control signal; When the power module receives the steering control signal and the steering angle control signal, it drives the drone to turn and adjusts the steering angle of the drone according to the steering angle control signal. When the power module receives the flight speed control signal, it adjusts the flight speed of the drone; The image acquisition module is used to collect images of the environment where the drone is located and images below the drone, and transmit the collected images to the digital image transmission communication module; The digital image transmission communication module is used to transmit the image information collected by the image acquisition module back to the ground station; The ground station processes the received image information, generates flight instructions, and transmits the flight instructions to the flight control module of the UAV; The positioning and navigation system is used to locate the object throwing point; After receiving the object delivery signal, the throwing device throws the object carried by the drone; The drone is equipped with a power module, an image acquisition module, a flight control module, a digital image transmission communication module, a positioning and navigation system, and a throwing device.

2. The drone with both detection and throwing functions according to claim 1, characterized in that: The drone is also equipped with a power module; The power supply module supplies power to the flight control module, the power module, the image acquisition module, the digital image transmission communication module, the positioning and navigation system, and the throwing device.

3. The drone with both detection and throwing functions according to claim 1, characterized in that: The image information collected by the image acquisition module is transmitted back by the digital image transmission communication module using the opencv digital recognition algorithm.

4. The drone with both detection and throwing functions according to claim 3, characterized in that: The steps of the opencv digital recognition algorithm include: 1) Convert the RGB information of the acquired image into the HSV color space to identify the color of the target; 2) Extract the minimum circumscribed rectangle of the red and blue color areas in the HSV color space and convert it into a binary image to obtain a rectangular binary image; 3) Store the rectangular binary image in the defined vector container; 4) Prepare template images of ten numbers from 0 to 9 and convert them into binary images to obtain template binary images; 5) Compare the rectangular binary image in the vector container with the template binary image to obtain the similarity; 6) Compare the similarity between each rectangular binary image and the template binary image, and the template image number corresponding to the highest similarity value is the recognized number.

5. The drone with both detection and throwing functions according to claim 1, characterized in that: Set the drone's default coordinate system to be consistent with the map coordinate system; The positioning and navigation system adopts a moving point comparison correction algorithm to perform deviation correction.

6. The drone with both detection and throwing functions according to claim 5, characterized in that: The steps of the moving point comparison correction algorithm include: 1) Set the take-off position of the drone to (x0, y0) and the target point position to (x i ,y i ), i is the target point ordinal number, i = 1, 2, ..., n, n is the total number of target points; 2) The drone flies to the target point (x i ,y i ), the actual position reached by the drone is (x′ i , y′ i ); 3) The recognition point of the drone for the target image is (x″ i , y″ i ); 4) Compare the actual position (x′ i , y′ i ) and drone identification point (x″ i , y″ i ) to correct the position deviation of the UAV.

Images