CN110562455B

CN110562455B - Intelligent unmanned aerial vehicle automatic charging system

Info

Publication number: CN110562455B
Application number: CN201910839143.4A
Authority: CN
Inventors: 刘耕远; 王文思
Original assignee: Beijing University of Technology
Current assignee: Beijing University of Technology
Priority date: 2019-09-05
Filing date: 2019-09-05
Publication date: 2021-04-30
Anticipated expiration: 2039-09-05
Also published as: CN110562455A

Abstract

The invention discloses an intelligent automatic charging system for unmanned aerial vehicles, including an unmanned aerial vehicle and a charging pile. module; the charging pile includes: a camera module, an FPGA development board module, an XY axis servo motor module, and a wireless charging module. The invention realizes that after the UAV landed on the UAV charging pile, the image of the lower part of the platform is collected by the image acquisition system, input into the FPGA, and the corresponding algorithm is used to calculate and recognize the characteristic mark of the lower part of the UAV, and its coordinates are calculated, Control the XY axis servo motor to move the wireless charging board to the bottom of the drone to realize automatic charging. It is suitable for automatic charging of various types of drones to achieve long-term battery life of the drone.

Description

Intelligent unmanned aerial vehicle automatic charging system

Technical Field

The invention relates to the technical field of unmanned aerial vehicles, in particular to an intelligent automatic charging system for an unmanned aerial vehicle.

Background

With the high-speed development of frontier defense, ship-borne and vehicle-mounted unmanned aerial vehicles, the unmanned aerial vehicles have obvious over-the-horizon, unmanned and networked trends, and have wide application in the field of national defense, power line patrol, disaster monitoring and the like. The problem that the duration is short is ubiquitous to present unmanned aerial vehicle, because the restriction in the battery technology, unmanned aerial vehicle's duration is general weak, generally only within 30 minutes. If need unmanned aerial vehicle timing or continuous operation, need manual operation and supervision to guarantee it at the course of the work at present, return to the air and descend to charge or change the battery, the manpower demand is big, and is with high costs to seriously influence work efficiency.

At present, unmanned aerial vehicles commonly used in production and life mainly divide into three kinds, fixed wing type flight wing, helicopter type aircraft and many rotor crafts, and many rotor crafts use four rotor crafts as main. Because the multi-rotor aircraft has low requirement on the takeoff field and the aircraft body is light, the multi-rotor aircraft can easily realize hovering and quickly change the course, and is widely applied to production and life. Typically, a complete set of four-rotor aircraft systems includes several subsystems, namely an aircraft system, a ground monitoring station, a communication link, and a payload system.

However, the endurance time of the existing four-rotor aircraft is about 20-30 minutes in the working process, and the endurance time of the aircraft is severely limited by the battery capacity. The common method is to increase the battery capacity, but the increase of the battery capacity causes the increase of the load of the aircraft, the output power rises along with the increase of the battery capacity, the endurance time increases along with the increase of the battery capacity, the manufacturing cost and the battery technology limit the endurance time of the aircraft, and the problem of long-time endurance of the aircraft cannot be fundamentally solved.

At present there has been unmanned aerial vehicle full-automatic ground charging platform abroad, because the air current that receives the change of environment atmospheric pressure wind direction and unmanned aerial vehicle production influences the accurate fixed point that makes unmanned aerial vehicle and descends and become very difficult, for solving above problem, adopt wireless charging technique more at present, adopt non-contact's mode to realize electric energy transmission, reduce the requirement to descending the accuracy.

Accordingly, this document presents an automatic wireless charging system for unmanned aerial vehicles. When the aircraft need charge, only need drop to the platform that charges, need not accurate descending, the servo motor system of platform below can remove the transmission section of wireless charger to unmanned aerial vehicle's below, charges for the aircraft group battery through wireless power transmission mode, and the back is accomplished in charging, and the aircraft returns and continues to carry out the work task.

Disclosure of Invention

Aiming at the defects in the prior art, the invention provides an intelligent automatic charging system of an unmanned aerial vehicle, which aims to solve the problems of poor universality, low efficiency and the like in the prior art.

The invention provides an intelligent unmanned aerial vehicle automatic charging system, which comprises an unmanned aerial vehicle and a charging pile, wherein the unmanned aerial vehicle comprises:

the low-power alarm module is used for monitoring the power of the battery in real time and outputting a signal of a charging request;

the unmanned aerial vehicle communication module is connected with the low-power alarm module and used for sending a charging request signal to the master control station through radio, and sending the position information to the positioning module by the unmanned aerial vehicle communication module after the master control station permits the charging request and sends the charging request to the unmanned aerial vehicle communication module;

the positioning module is connected with the unmanned aerial vehicle communication module and used for navigation and positioning of the unmanned aerial vehicle;

the flight control module is connected with the positioning module and used for controlling the unmanned aerial vehicle to land to a charging pile position which can be used by the general control platform;

the charging butt joint module comprises a wireless coil, is placed below the unmanned aerial vehicle and is used for butt joint with the charging pile wireless charging panel module;

fill electric pile includes:

and the camera module is used for acquiring an image of the position of the aircraft on the platform in the docking process of the aircraft and the charging station.

The FPGA development board module is used for the camera module to acquire an image of the unmanned aerial vehicle parked at the position of the unmanned aerial vehicle charging pile, the image is input into the FPGA development board, the characteristic mark of the lower part of the unmanned aerial vehicle is calculated and identified by utilizing a corresponding algorithm, the coordinate of the characteristic mark is calculated, the XY-axis servo motor is controlled to move the wireless charging board to the lower part of the unmanned aerial vehicle, and automatic charging is realized.

And the XY-axis servo motor module is used for controlling the wireless charging module to move after receiving the control signal sent by the FPGA development board module, so that the module can freely move below the charging platform, and when the aircraft lands on the charging platform, the wireless charger is moved to a position right below a target, so that the automatic charger of the aircraft is realized.

The wireless charging module is used for being in butt joint charging with a charging butt joint module arranged below the unmanned aerial vehicle.

As a further improvement of the invention, the positioning module comprises an indoor UWB accurate positioning unit and a Beidou positioning unit.

As a further improvement of the present invention, the charging docking module includes a wireless coil, and the wireless charging module includes a Baseus wireless charger, so as to increase the receiving area of the wireless coil to increase the charging speed.

As a further improvement of the invention, the XY-axis servo motor module comprises a cross module sliding table (XY-axis sliding table), so that the driving capability of the cross module sliding table is improved, and the moving speed is increased.

The invention provides an efficient intelligent unmanned aerial vehicle automatic charging system, which realizes that an unmanned aerial vehicle is landed on an unmanned aerial vehicle charging pile, then an image at the lower part of a platform is acquired through an image acquisition system, the image is input into an FPGA (field programmable gate array), a characteristic mark at the lower part of the unmanned aerial vehicle is calculated and identified by using a corresponding algorithm, the coordinate of the characteristic mark is calculated, and an XY-axis servo motor is controlled to move a wireless charging plate to the lower part of the unmanned aerial vehicle, so that automatic charging. Be applicable to and carry out automatic charging to the unmanned aerial vehicle of multiple model to realize that unmanned aerial vehicle lasts for a long time.

Drawings

Fig. 1 is a schematic structural diagram of an unmanned aerial vehicle automatic charging pile system provided by the invention.

Fig. 2 is a schematic diagram of a module structure of the unmanned aerial vehicle system provided by the invention.

Fig. 3 is a schematic structural diagram of a module of the automatic charging pile of the unmanned aerial vehicle provided by the invention.

Fig. 4 is a schematic view of connection between a camera module and a display module of the automatic charging pile of the unmanned aerial vehicle provided by the invention.

Fig. 5 is a schematic technical implementation diagram of the automatic unmanned aerial vehicle charging pile provided by the invention.

Detailed Description

The present invention will be described in further detail with reference to the accompanying drawings and specific embodiments.

As shown in fig. 1, 2 and 3, the invention discloses an intelligent unmanned aerial vehicle automatic charging system, which comprises an unmanned aerial vehicle 1 and a charging pile 7, wherein the unmanned aerial vehicle 1 is used for matching with the charging pile 7 to complete charging work. A wireless coil is arranged below the unmanned aerial vehicle 1; be equipped with wireless charging panel on the stake of charging 7, wireless charging panel and wireless coil interact accomplish and charge. Fill the wireless platform that charges that electric pile 7 adopted and be the translucent platform of ya keli material, upwards have an image acquisition device from the bottom, gather platform bottom image in real time to carry out image identification. The lower part of the unmanned aerial vehicle is provided with a wireless charging coil for receiving electric quantity. After unmanned aerial vehicle 1 descends to filling the electric pile platform, fill electric pile system discernment target, system control motor system removes wireless charging base to the below of wireless charging coil, carries out wireless charging.

The unmanned aerial vehicle 1 comprises a low-power alarm module 2, a communication module 3, a positioning module 4, a flight control module 5 and a charging butt-joint module 6;

the low-power alarm module 2 is used for detecting the power of the battery and outputting a signal indicating whether charging is needed or not, the low-power alarm module 2 is connected with the communication module 3, and the battery of the unmanned aerial vehicle is placed in the unmanned aerial vehicle and connected with the wireless coil to obtain the power charged by the wireless coil;

the communication module 3 is connected with the low-power alarm module 2, after receiving an electric signal sent by the low-power alarm module 2, the communication module 3 sends a charging request signal to a master control station through radio, after the master control station permits, the communication module 3 sends charging pile position information to the unmanned aerial vehicle communication module 3, the communication module 3 sends the charging pile 7 position information to the positioning module 4 in an electric signal mode, the positioning module 4 sends a position information signal to the flight control module 5, and the flight control module 5 controls the unmanned aerial vehicle 1 to land at a target position;

the positioning module 4 is connected with the flight control module 5, the navigation positioning of the unmanned aerial vehicle 1 is realized through the positioning module 4, the positioning module comprises an indoor UWB (ultra wide band) precise positioning unit and a Beidou positioning unit, the UWB precise positioning unit carries out real-time position positioning, the positioning precision can be in a sub-meter level, and the Beidou positioning unit can realize the global positioning function;

flight control module 5 with orientation module 4 is connected, and flight control module 5 is used for controlling unmanned aerial vehicle 1 to descend to the target location.

The charging docking module 6 is arranged at the bottom of the unmanned aerial vehicle 1 and comprises a wireless coil, and the charging docking module 6 is used for docking with the charging pile wireless charging module 11;

fill electric pile 7 includes: the system comprises a camera module 8, an FPGA development board module 9, an XY axis servo motor module 10, a display module 11 and a wireless charging module 12;

the camera module 8 comprises a CMOS chip image sensor OV5640 camera, is connected with the FPGA development board module 9 to realize image transmission, and is used for acquiring images of the unmanned aerial vehicle 1 on the platform in the butt joint process of the unmanned aerial vehicle 1 and the charging pile 7. When unmanned aerial vehicle 1 descends to charging pile 7, the camera module needs upwards gather the image in charging platform bottom, carries out image recognition, detects the position information that the target aircraft descends to charging platform.

The FPGA development board module 9 comprises a SPARTAN6 series FPGA development board of XILINX company, the model is XC6SLX9-2FTG256C, and is connected with the camera module 8, the XY axis servo motor module 10 and the wireless charging module 11, and the FPGA development board module 9 is connected with the camera module 8 through a DVP interface to realize image transmission and input the image transmission into the FPGA development board. And performing image recognition based on the obtained ideal binary image to obtain the accurate position of the target on the platform, and controlling the XY-axis servo motor module 10 to move the wireless charging plate to the lower part of the unmanned aerial vehicle to realize automatic charging.

And the display module 11 is connected with the FPGA development board module 9, so that the visualization function of the system can be realized. The image display device is used for displaying the image acquired by the camera module in real time, and the process of carrying out binarization on the target is also displayed on the image display device. The display module can also adjust the position of the camera.

According to the register setting of the camera module, the image input to the FPGA development board by the camera module is a 16-bit serial signal with the format of RGB565, and finally the image needs to be converted into the format of YUV 888. The following steps are required.

S1, splitting an RGB565 signal into three parts, namely RED-5, GREEN-6 and BLUE-5; according to the format, the 16-bit image signal has blue color signals from 0 th bit to 4 th bit, green color signal lines from 5 th bit to 10 th bit, and red color signal lines from 11 th bit to 15 th bit.

S2, respectively expanding the split red, green and blue signals into 8 bits;

according to the definition of RGB565 to RGB888, only 3 bits and 0 need to be added after the red 5-bit signal, and similarly, 2 and 30 need to be added after the green 6-bit signal and the blue 5-bit signal, respectively. The value range of the 8-bit binary signal is 0-255, and the value range of the last 3-bit signal is 0-7, so that the expansion into 8 bits has little influence on the image color.

S3, conversion is completed through a formula of converting RGB into YUV;

and S4, finally, arranging and combining three 8 bits according to the Y.U.V sequence to form a 24-bit image signal, and outputting a 24-bit serial image signal with V concentration signals from the 0 th bit to the 7 th bit, U chrominance signals from the 8 th bit to the 15 th bit and Y brightness signals from the 16 th bit to the 23 th bit.

An image edge refers to the collection of pixels around an image where the gray level of the pixels changes sharply, which is the most fundamental feature of the image. The Sobel operator uses a first order differential operator to detect the edge of the image, and the first order differential edge operator (gradient edge operator) uses the step property of the image at the edge to detect the edge. In the actual algorithm implementation process, a 3 x 3 template is used as a core to perform convolution calculation with each pixel point of a selected area with the same size in an image, gradient values of the selected image in the horizontal direction and the vertical direction are calculated, and the horizontal gray value and the longitudinal gray value of each pixel point of the image are combined through a corresponding formula to obtain the gradient value of the point in one direction. Since the classical Sobel algorithm is sensitive to noise, the threshold needs to be determined, and misjudgment of the edge point is often caused. To solve this problem, there are many adaptive threshold selection methods, such as histogram method. However, the histogram method requires that the gray histogram of the image after the edge detection by the Sobel algorithm presents obvious double peaks, and the binarization is realized by selecting the gray level corresponding to the valley bottom between the two peaks as the threshold value, so that the selection of the adaptive threshold value by the histogram method is greatly limited. Based on the method, after the image is subjected to convolution gradient value calculation, the median filtering and weighted average method is adopted to realize the calculation of the self-adaptive threshold value so as to realize the binarization of the image.

Because the weighted gray level of the central pixel corresponds to the gray level of the weighted central pixel after the Sobel operator is subjected to gradient transformation, the gray levels of the pixel points at the edge and the pixel points at the non-edge have larger difference, the traditional method is to try to set a fixed threshold value to realize binarization, so the processing is simple but seriously influenced by external factors, in order to realize threshold value self-adaptive adjustment, a 5 x 5 region is selected from an image after convolution operation, and the median value, the secondary minimum value and the minimum value of the region are calculated by using median filtering. And carrying out weighted average on the 3 values, taking the result as a threshold value, and reducing the selection of the weight value along with the increase of the distance from the central pixel point.

And carrying out binarization processing on the image according to the gray level of the image and the data obtained by the edge detection processing.

Firstly, according to the characteristics of the red LED small lamp which needs to be identified, the obtained target is much brighter than the background of the view field, and the color is red and is more remarkable. Therefore, a predetermined threshold value is set for the brightness, the chromaticity, and the density, which are three parameters Y, U, V, respectively, based on the obtained image YUV data. And when the values of the three parameters in one pixel block are greater than the set threshold value, the pixel block is determined as a target pixel block, and if one of the three parameters does not meet the requirement, the pixel block is determined not to meet the requirement and is not the target pixel block. According to the actual image and the matlab simulation result, when the brightness is greater than 250, the concentration is greater than 210 and the chroma is greater than 200, more than 98% of interference noise can be removed on the premise of extracting the target image.

Then, the result obtained by the self-adaptive threshold edge detection is utilized to process the image,

and performing cross comparison on the results of the two image processing, determining that the results are correct judgment when the results of the two image processing are consistent, and determining that the results obtained by default image gray scale processing are correct judgment when the results of the two image processing are inconsistent.

And the image is stored by writing an FIFO program by using the FPGA. Because the image data volume is large, the storage of one frame of image data cannot be completed inside the FPGA, and therefore the dual-port RAM is generated for caching by using the inherent IP core on the FPGA. In order to solve the problem of image smear, a ping-pong structure is adopted to read and write image data. The SDRAM on the development board is divided into two areas, namely a Bank0 area and a Bank1 area, and when data collected by the camera is written into a Bank0 area of the SDRAM, the image processing module reads the data from the Bank1 area of the SDRAM to process the data. When one-frame image writing of the camera is completed, Bank exchange of SDRAM reading and writing is performed.

In order to extract the target coordinates, the features of the target need to be analyzed. In the picture where binarization has been completed, most of the background, i.e. white, is targeted to a black solid circle, meaning that there will be blocks of pixels of consecutive rows and columns appearing black, and the following algorithm is designed.

And establishing a rectangular coordinate system by taking the upper left corner of the LCD screen as an original point, and traversing one frame of image by one pixel point. Two signals are defined, a row-sequential signal and a column-sequential signal, respectively, for indicating whether or not to traverse to consecutive black pixel blocks.

The XY-axis servo motor module 10 comprises an XY stepping motor sliding table group and is used for determining position information of the unmanned aerial vehicle 1 on an XY axis of the charging pile 7 through an image recognition module 8 after the unmanned aerial vehicle 1 lands on the unmanned aerial vehicle automatic charging pile 7, transmitting the position information to an FPGA development board module 9 through an electric signal, sending the electric signal to control the FPGA development board module 9 to move a wireless charging module 12 below the unmanned aerial vehicle 1, and sending the electric signal to the FPGA development board module 9;

the low-power alarm module 2 can detect the power utilization condition of the unmanned aerial vehicle 1, when the power is too low, a signal needing to be charged is output to the communication module 3, the communication module 3 sends a signal to the general control station through radio to request charging, the general control station can send position information to the charging pile 7 to the communication module 3 after allowing the charging request, the communication module 3 sends the position information to the positioning module 4, the positioning module 4 positions the position and sends the signal to the flight control module 5, and the flight control module 5 controls the unmanned aerial vehicle to search the position of the charging pile according to the unmanned aerial vehicle positioning module 4 and control the charging pile to fly to a specified place; when arriving the position, unmanned aerial vehicle 1 berths at the translucent platform that loads the ya keli material, and unmanned aerial vehicle 1 obtains the image of unmanned aerial vehicle 1 position on the platform with filling 7 butt joints in-process of electric pile. When the unmanned aerial vehicle 1 lands on the charging pile 7, the camera module needs to upwards collect images at the bottom of the charging platform to perform image recognition, the position of the target aircraft landing on the charging platform is detected, the camera module 8 is connected with the FPGA development board module 9 through a DVP interface to realize image transmission, the images are input into the FPGA development board, the results obtained by the image processing part and the self-adaptive threshold image binarization and the target characteristic value binarization are compared by utilizing the characteristics of the image processing part and the self-adaptive threshold image binarization and the target characteristic value binarization, most of interference noise points are eliminated, and a final binarization image is obtained. And carrying out image recognition based on the obtained ideal binary image to obtain the accurate position of the target on the platform, and controlling the XY-axis servo motor module 10 to move the wireless charging module 12 to the lower part of the unmanned aerial vehicle to realize automatic charging.

Preferably, the positioning module 4 comprises a Beidou positioning unit and a UWB positioning unit. Through the mutual cooperation of indoor accurate positioning unit of UWB and big dipper positioning unit, compound navigation carries out, realizes accurate location, and sub-meter level can be accomplished to positioning accuracy, ensures that unmanned aerial vehicle 1 stops in reasonable position, reduces the error.

Preferably, wireless charging module 12, including wireless charging panel for the wireless coil butt joint of placing with the unmanned aerial vehicle below charges, and its efficiency is greater than wired charging, and the convenience is high moreover.

Preferably, XY axle servo motor module 10 is used for removing wireless charging panel to the unmanned aerial vehicle below after 1 landing of unmanned aerial vehicle through image recognition confirm the XY epaxial positional information of unmanned aerial vehicle on filling electric pile, and this servo motor system comprises direct current 5V power, direct current 24V power, controller, drive plate, driver, XY step motor slip table group. The controller can realize the functions of manually controlling the four-way movement and manual reset of the sliding table and the like, and is connected with the driving plate through 8pin flat cables; the driving plate is a booster circuit which can boost the +5V or +3.3V voltage output by the controller to 24V and is connected to the driver through a copper wire; the driver is a driving device of the stepping motor, can change the dynamic current state, half-current or full-current state, rotating speed and other parameters of the stepping motor by changing the arrangement and combination of 8 toggle switches of the driver, and is connected with the XY stepping motor sliding table group through a copper wire; the XY stepping motor sliding table group is composed of mutually vertical and same sliding table groups, a black platform is arranged on the Y-axis sliding table and can change positions along with the movement of the XY stepping motor sliding table group, and the working range of the platform is a horizontal square of 50CM x 50 CM. The XY axis servo motor module 10 has the characteristics of high response speed, high moving speed, small error and the like, and meanwhile, in order to ensure the stability of the system, the structure of the servo motor module cannot be complex and can work for a long time.

The foregoing is a more detailed description of the invention in connection with specific preferred embodiments and it is not intended that the invention be limited to these specific details. For those skilled in the art to which the invention pertains, several simple deductions or substitutions can be made without departing from the spirit of the invention, and all shall be considered as belonging to the protection scope of the invention.

Claims

1. An intelligent automatic charging system for unmanned aerial vehicles, comprising an unmanned aerial vehicle and a charging pile, and the unmanned aerial vehicle comprises:

The low battery alarm module is used to monitor the battery power in real time and output the signal of charging request;

The drone communication module is connected to the low battery alarm module, and is used to send a charging request signal to the main console by radio. When the main console approves the charging request and sends the location information to the drone communication module, the drone The communication module sends the position information to the positioning module;

The positioning module is connected with the UAV communication module for navigation and positioning of the UAV;

The flight control module is connected with the positioning module, and is used to control the drone to land to the position of the charging pile that can be arranged by the main console;

The charging docking module includes a wireless coil and is placed under the drone for docking with the charging pile wireless charging board module;

The charging pile includes: a camera module for acquiring an image of the position of the aircraft on the platform during the docking process of the aircraft and the charging station;

The FPGA development board module is used for the camera module to obtain the image of the position where the drone is parked on the charging pile of the drone, input it into the FPGA development board, and use the corresponding algorithm to calculate and identify the characteristic signs of the lower part of the drone, and calculate its value. Coordinates, control the XY axis servo motor to move the wireless charging board to the bottom of the drone to realize automatic charging;

The XY axis servo motor module is used to control the movement of the wireless charging module after receiving the control signal from the FPGA development board module, so that the module can move freely under the charging platform. When the aircraft lands on the charging platform, the wireless charging Move directly below the target to realize the automatic charger of the aircraft;

The wireless charging module is used for docking charging with the charging docking module placed under the drone;

According to the register settings of the camera module, the image input by the camera module to the FPGA development board is a 16-bit serial signal in RGB565 format, and finally it needs to be converted into YUV888 format; the following steps are required;

S1. Divide the RGB565 signal into three parts: RED-5, GREEN-6, and BLUE-5; according to the name of the format, it can be known that the 0th to 4th bits of this 16-bit image signal are blue signals , the 5th to 10th bits are green signal lines, and the 11th to 15th bits are red signal lines;

S2. Expand the split red, green and blue signals into 8 bits respectively;

According to the definition of RGB565 to RGB888, we only need to add 3 bits of 0 after the red 5-bit signal. Similarly, add 2 and 3 0s after the green 6-bit signal and the blue 5-bit signal respectively; 8-bit binary The signal is 0-255, and the value range of the last 3-bit signal is 0-7, so expanding to 8 bits has little effect on the image color;

S3. Complete the conversion through the formula of RGB to YUV;

S4. Finally, the three 8-bits are arranged and combined into a 24-bit image signal in the order of Y.U.V. The output 24-bit serial image signal The 0th to 7th bits are V density signals, and the 8th to 15th bits are U color. brightness signal, the 16th to 23rd bits are the Y brightness signal.

2 . An intelligent automatic charging system for unmanned aerial vehicles according to claim 1 , wherein the positioning module comprises an indoor UWB precise positioning unit and a Beidou positioning unit. 3 .

3 . The intelligent automatic charging system for unmanned aerial vehicles according to claim 1 , wherein the charging docking module comprises a wireless coil, and the wireless charging module comprises a Baseus wireless charger. 4 .

4. An intelligent automatic charging system for unmanned aerial vehicles according to claim 1, characterized in that: the XY axis servo motor module comprises a cross module slide table, and the driving capability of the cross module slide table is improved to improve the moving speed .

5. A kind of intelligent automatic charging system for unmanned aerial vehicle according to claim 1, it is characterized in that: camera module, comprises CMOS chip image sensor OV5640 camera, is connected with FPGA development board module to realize image transmission, is used for unmanned aerial vehicle The image of the position of the UAV on the platform is obtained during the docking process with the charging pile; when the UAV lands on the charging pile, the camera module needs to collect images upward at the bottom of the charging platform, perform image recognition, and detect the landing of the target aircraft on the charging platform. location information on .

6. a kind of intelligent unmanned aerial vehicle automatic charging system according to claim 1, is characterized in that: FPGA development board module, comprises the SPARTAN6 series FPGA development board of XILINX company, model is XC6SLX9-2FTG256C, and camera module, XY The axis servo motor module and the wireless charging module are connected for the connection of the camera module with the FPGA development board module through the DVP interface to realize image transmission, input into the FPGA development board, and use the image processing part to combine the adaptive threshold image binarization and target eigenvalues The characteristics of binarization, compare the results obtained by the two, eliminate most of the interference noise, and obtain the final binarized image; carry out image recognition based on the obtained ideal binarized image, and obtain the target on the platform. Accurate position, and control the XY axis servo motor module to move the wireless charging board to the bottom of the drone to achieve automatic charging.

7. An intelligent automatic charging system for unmanned aerial vehicles according to claim 1, characterized in that: a display module, the display module is connected to an FPGA development board module to realize the visualization function of the system; it is used to display the image collected in real time by the camera module , the process of binarizing the target is also shown above; the display module can also adjust the position of the camera.