Unmanned aerial vehicle secondary development teaching system and identification method

Abstract

The invention relates to the technical field of unmanned aerial vehicles, and discloses an unmanned aerial vehicle secondary development teaching system and an identification method, and the system comprises an unmanned aerial vehicle which is provided with a holder and a high-definition camera, the holder is used for adjusting the shooting angle of the high-definition camera, and the high-definition camera is used for capturing a real-time image; the ground mark is used for guiding the unmanned aerial vehicle to track and fly; the commodity shelf is used for placing to-be-identified articles of the unmanned aerial vehicle; the ground station is in communication connection with the unmanned aerial vehicle and is used for receiving the images and data transmitted by the unmanned aerial vehicle and processing and analyzing the images and data; and the sensor assembly is arranged on the unmanned aerial vehicle and is used for realizing environment sensing and positioning of the unmanned aerial vehicle. According to the invention, accurate positioning and sensing of the unmanned aerial vehicle in a complex environment are realized through the unmanned aerial vehicle integrated with the holder and the high-definition camera, the ground mark for guiding tracking, the object placement rack for placing the to-be-identified object, the ground station for receiving and analyzing data, and the sensor assembly equipped with the binocular vision sensor, the ranging radar and the infrared sensor.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned aerial vehicles, and in particular to a secondary development teaching system and an identification method for unmanned aerial vehicles. Background Art

[0002] In current drone applications, drones are integrated with artificial intelligence technology, equipped with a variety of sensors, and integrated with programming languages, single-chip microcomputers, image recognition and other technologies to control drones to complete work tasks in certain scenarios, which has become a hot topic in current drone applications.

[0003] However, there is no dedicated UAV secondary development teaching system on the market. In the traditional teaching system, the teaching of UAV technology is often limited to theoretical knowledge explanation and simple simulated flight training. This teaching method makes it difficult for students to fully understand and master the complex operation and autonomous flight capabilities of UAVs in practical applications.

[0004] In addition, most existing drone teaching systems lack teaching resources that are closely integrated with actual application scenarios, such as racks, ground signs and lines, which makes it difficult for students to combine theoretical knowledge with practical operations during the learning process, limiting the development of their practical ability and innovative thinking. Summary of the invention

[0005] In view of the deficiencies in the prior art, the present invention proposes a UAV secondary development teaching system and an identification method to solve the above technical problems.

[0006] In the first aspect, a UAV secondary development teaching system is provided, including:

[0007] The drone is equipped with a gimbal and a high-definition camera. The gimbal is used to adjust the shooting angle of the high-definition camera, and the high-definition camera is used to capture real-time images.

[0008] Ground signs, used to guide the drone to follow the track;

[0009] The storage rack is used to place items to be identified by the drone;

[0010] The ground station is connected to the drone for receiving, processing and analyzing the images and data transmitted by the drone;

[0011] The sensor component is installed on the drone to realize the drone's environmental perception and positioning.

[0012] Furthermore, the sensor assembly includes a binocular vision sensor, a ranging radar and an infrared sensor.

[0013] Furthermore, the drone also includes an accelerometer, a gyroscope and a magnetometer for determining the flight direction and attitude of the drone.

[0014] Furthermore, the ground station is provided with a model training tool and an image recognition tool for performing model training and recognition processing on images taken by the drone.

[0015] In a second aspect, a recognition method for a UAV secondary development teaching system is provided, comprising the following steps:

[0016] When the drone follows the ground mark, it obtains the target image of the rack area through the high-definition camera;

[0017] The drone performs target recognition on the target image to obtain location information of the target object;

[0018] The UAV adjusts the flight attitude of the UAV according to the position information of the target object;

[0019] The drone sends the target image to the ground station.

[0020] Furthermore, the drone performs target recognition on the target image to obtain the location information of the target object, including:

[0021] Convert the color image to obtain a grayscale image;

[0022] Performing Gaussian filtering and smoothing processing on the grayscale image to obtain a smoothed grayscale image;

[0023] Performing edge detection on the smoothed grayscale image to obtain edge information of the image;

[0024] Post-processing and Hough transform algorithm are used to extract edge information and determine the location information of the target object.

[0025] Furthermore, the method further comprises:

[0026] The drone identifies the charging area marker in the charging area;

[0027] By calculating the relative position between the gimbal and the center point of the charging area marker, the deviation between the current position of the drone and the charging area is obtained;

[0028] The flight attitude and position of the drone are adjusted according to the deviation between the current position of the drone and the charging area to obtain the fixed landing position.

[0029] Furthermore, the method further comprises:

[0030] The target image transmitted by the UAV is used to train the landmark recognition model to obtain the landmark recognition model;

[0031] Using the landmark recognition model to perform landmark recognition on the newly input image to obtain a recognition result;

[0032] The recognition results are sorted to obtain a recognition result statistics table.

[0033] The invention adopting the above technical solution has the following advantages:

[0034] The present invention realizes the precise positioning and perception of the UAV in a complex environment through a UAV with an integrated pan-tilt platform and a high-definition camera, ground markings to guide tracking, a shelf to place items to be identified, a ground station to receive and analyze data, and a sensor component equipped with a binocular vision sensor, a ranging radar and an infrared sensor. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] In order to more clearly illustrate the specific implementation of the present invention, the following will briefly introduce the drawings required for use in the specific implementation. In all the drawings, each element or part is not necessarily drawn according to the actual scale.

[0036] Figure 1 A schematic diagram of a UAV secondary development teaching system of the present invention;

[0037] Figure 2 A schematic diagram of a storage rack in a secondary development teaching system for a drone of the present invention;

[0038] FIG3 is a schematic diagram of an image captured in a UAV secondary development teaching system of the present invention;

[0039] FIG4 is a schematic diagram of image edge information in a UAV secondary development teaching system of the present invention (I);

[0040] Figure 5 This is a schematic diagram of image edge information in a UAV secondary development teaching system of the present invention (II);

[0041] Figure 6 A schematic diagram of a plane rectangular coordinate system in a UAV secondary development teaching system of the present invention;

[0042] Figure 7 A schematic diagram of a charging area in a UAV secondary development teaching system of the present invention;

[0043] Figure 8 The present invention is a flow chart of an identification method for a secondary development teaching system of an unmanned aerial vehicle. DETAILED DESCRIPTION

[0044] The following embodiments of the technical solution of the present invention are described in detail in conjunction with the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solution of the present invention, and are therefore only used as examples, and cannot be used to limit the protection scope of the present invention.

[0045] like Figures 1 to 8 As shown, a UAV secondary development teaching system of the present invention includes:

[0046] The drone is equipped with a gimbal and a high-definition camera. The gimbal is used to adjust the shooting angle of the high-definition camera, and the high-definition camera is used to capture real-time images.

[0047] Ground signs, used to guide the drone to follow the track;

[0048] The shelf is used to place items to be identified by the drone (several cold water cups and hot water cups);

[0049] The ground station is connected to the drone via WIFI communication and is used to receive images and data transmitted by the drone and process and analyze them;

[0050] The sensor component is installed on the drone to realize the drone's environmental perception and positioning.

[0051] Specifically, the drone has integrated a gimbal and a high-definition camera, ground landmarks to guide tracking, shelves to place items to be identified, a ground station to receive and analyze data, and a sensor component equipped with a binocular vision sensor, ranging radar and infrared sensor, to achieve precise positioning and perception of the drone in complex environments.

[0052] In some embodiments, the sensor assembly includes a binocular vision sensor, a range-finding radar, and an infrared sensor;

[0053] In some embodiments, the drone further includes an accelerometer, a gyroscope, and a magnetometer for determining the flight direction and attitude of the drone;

[0054] In some embodiments, the ground station is provided with a model training tool and an image recognition tool for performing model training and recognition processing on images taken by the drone.

[0055] Specifically, the UAV has a wheelbase of 360mm, and the fuselage and propellers are designed as an integrated whole to avoid flight jitter caused by the subsequent installation of propellers.

[0056] The front bottom of the drone is equipped with a high-definition camera through a gimbal, and the gimbal's rotation angle range is: -90°-20°; the drone's frame has customized sensor mounting brackets and mounting holes at four positions, front, back, left, and right, to facilitate students to install specific sensors according to task requirements.

[0057] The drone is equipped with an accelerometer, gyroscope, and magnetometer, and the flight controller controls the direction and distance of the drone. The drone has a lightweight body design to improve its endurance.

[0058] The drone is equipped with four binocular vision sensors, four ranging radars, and one infrared sensor. Sensor plug-in holes are reserved on the central board, and students can choose to install sensors according to actual needs.

[0059] There are several cardboard boxes on the shelf, with patterns of different colors and shapes posted on the front of the boxes; there are also several water cups that can withstand temperatures of 100°C.

[0060] In other embodiments, a recognition method for a drone secondary development teaching system is provided, comprising the following steps:

[0061] Step S01: When the drone is tracking according to the ground mark, a target image of the rack area is obtained through a high-definition camera;

[0062] Step S02: The drone performs target recognition on the target image to obtain the location information of the target object;

[0063] Step S03: the UAV adjusts its flight attitude according to the position information of the target object;

[0064] Step S04: The UAV sends the target image to the ground station.

[0065] In some embodiments, the drone performs target recognition on the target image to obtain the location information of the target object, including:

[0066] Convert the color image to obtain a grayscale image;

[0067] Performing Gaussian filtering and smoothing processing on the grayscale image to obtain a smoothed grayscale image;

[0068] Perform edge detection on the smoothed grayscale image to obtain edge information of the image;

[0069] Post-processing and Hough transform algorithm are used to extract edge information and determine the location information of the target object.

[0070] Specifically, a high-definition camera is used to capture real-time images, and the input color image is converted into a grayscale image through image processing hardware. This conversion process not only simplifies the subsequent image processing steps, but also effectively reduces the computational complexity.

[0071] Use a Gaussian filter to smooth the grayscale image, reduce the noise and unnecessary details in the image, select the Canny edge detector for edge detection, and make the detected edges appear in red

[0072] Post-processing is used to connect broken edges or remove small noise edges, and finally edge analysis is performed in combination with Hough transform to determine the trajectory that the drone needs to track.

[0073] While the drone is flying, it captures clear images below in real time and analyzes them in detail through sophisticated image processing technology. A plane rectangular coordinate system with only the first quadrant is generated on each frame of the image, with the bottom edge of the image as the x-axis and the left edge as the y-axis. (The Y-axis is consistent with the direction of the nose, and the X-axis is perpendicular to the direction of the nose)

[0074] The system accurately locates the center point of the gimbal (x_target, y_target) and compares it with the preset tracking line to determine the deviation of the current flight path. Once the gimbal center point deviates from the ideal range of the tracking line, the drone immediately starts the adaptive adjustment mechanism.

[0075] The drone uses the gimbal center point as a reference, draws a straight line X perpendicular to the direction of the nose, and measures the angle A formed when this straight line intersects with the edge of the tracking line. Using this angle, the system quickly calculates the angle that the drone needs to spin by formula (90–A) to correct the direction deviation. If the calculation result is a positive number, the drone will spin counterclockwise; otherwise, it will spin clockwise.

[0076] At the same time, in order to adjust the drone's position in the horizontal direction (X-axis direction), the system also calculates the difference between the intersection point (x, y) and the gimbal center point (x_target, y_target), and converts this distance into the actual distance in meters through the formula (x-x_target)*0.001. Based on this calculation result, the drone can accurately determine which direction it should move on the X-axis to return to the tracking line: if the result is positive, the drone will fly to the right for the corresponding distance; if it is negative, it will fly to the left until the gimbal center point returns to the tracking line range.

[0077] Fixed-altitude flight achieves altitude adjustment control through the altitude sensor installed at the bottom of the drone.

[0078] In some embodiments, the method further comprises:

[0079] The drone identifies the charging area marker in the charging area;

[0080] By calculating the relative position between the gimbal and the center point of the charging area marker, the deviation between the current position of the drone and the charging area is obtained;

[0081] The flight attitude and position of the drone are adjusted according to the deviation between the current position of the drone and the charging area to obtain the fixed landing position.

[0082] Specifically, after the drone moves to the top of the charging area according to the flight route, it turns the drone gimbal to search for the charging area marker and takes a photo of the charging area. Based on the photo, the drone calculates the relative position of the gimbal and the center point of the charging area marker, and calculates the distance the drone needs to move to directly above the center of the red marker.

[0083] After the gimbal captures the marker location information map, the system will start a series of sophisticated image processing and analysis steps to generate a plane rectangular coordinate system with only the first quadrant on the image, with the bottom edge of the image as the x-axis and the left edge as the y-axis. (The Y-axis is consistent with the direction of the aircraft head, and the X-axis is perpendicular to the direction of the aircraft head)

[0084] First, through advanced color recognition technology, the system can accurately identify the markers in the image and further analyze their location characteristics. Next, the system calibrates the center point coordinates (x, y) of the charging area marker according to the coordinate system, and determines the center of the location information map and the center position coordinates of the drone gimbal (x_target, y_target). In this step, in order to facilitate observation and analysis, the system will use a green box on the image to mark the center position of the gimbal, and use a circular marker to highlight the center point of the charging area marker.

[0085] The distance from the center of the gimbal to the center of the drone in the Y direction is measured as y_0 (i.e., the Y-direction offset distance of the gimbal relative to the center of the drone). According to the structural characteristics of the drone itself, the center of the gimbal coincides with the center of the drone in the X direction, i.e., the distance is 0.

[0086] The system then uses a specific mathematical formula to calculate the relative position between the marker and the center of the drone. Specifically, through the formula (x-x_target)*0.001, the system can calculate the distance between the marker and the center of the drone in the lateral (or X-direction) direction, and convert this distance into the actual distance in meters. If the calculation result is greater than zero, it means that the marker is to the right of the center of the drone, and the system will instruct the drone to fly to the right; conversely, if the calculation result is less than zero, the marker is to the left of the center of the drone, and the system will instruct the drone to fly to the left.

[0087] The distance between the marker and the center of the drone in the longitudinal direction (or Y direction) is calculated using the formula (y-y_target+y_0)*0.001. This calculation result will also be converted into the actual distance in meters. If the calculation result is greater than zero, it means that the marker is in front of the center of the drone, and the system will command the drone to fly forward; conversely, if the calculation result is less than zero, the marker is behind the center of the drone, and the system will command the drone to fly backward.

[0088] Through accurate formula calculation, the system can accurately obtain the relative position relationship between the marker and the center of the drone gimbal, and issue precise flight control instructions based on this, ensuring that the drone can accurately locate the marker.

[0089] The drone moves to the top of the red marker and starts to land. After landing to a certain height (1.5m), the drone gimbal compares the relative position with the center of the red marker again, and then adjusts the position of the drone again according to the set target position. This step can be repeated multiple times according to the actual scene needs; then the final landing procedure is executed.

[0090] In some embodiments, the method further comprises:

[0091] The target image transmitted by the UAV is used to train the landmark recognition model to obtain the landmark recognition model;

[0092] Using the landmark recognition model to perform landmark recognition on the newly input image to obtain a recognition result;

[0093] The recognition results are sorted to obtain a recognition result statistics table.

[0094] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or replace some or all of the technical features therein by equivalents. These modifications or replacements do not deviate the essence of the corresponding technical solutions from the scope of the technical solutions of the embodiments of the present invention, and they should all be included in the scope of the claims and specification of the present invention.

Claims

1. A UAV secondary development teaching system, characterized in that: include: The drone is equipped with a gimbal and a high-definition camera. The gimbal is used to adjust the shooting angle of the high-definition camera, and the high-definition camera is used to capture real-time images. Ground signs, used to guide the drone to follow the track; The storage rack is used to place items to be identified by the drone; The ground station is connected to the drone for receiving, processing and analyzing the images and data transmitted by the drone; The sensor component is installed on the drone to realize the drone's environmental perception and positioning.

2. The system according to claim 1, characterized in that The sensor assembly includes a binocular vision sensor, a ranging radar and an infrared sensor.

3. The system according to claim 1, characterized in that The drone also includes an accelerometer, a gyroscope and a magnetometer for determining the flight direction and attitude of the drone.

4. The system according to claim 1, characterized in that The ground station is provided with a model training tool and an image recognition tool for performing model training and recognition processing on the images taken by the drone.

5. A recognition method for a UAV secondary development teaching system, characterized in that: The following steps are involved: When the drone follows the ground mark, it obtains the target image of the rack area through the high-definition camera; The drone performs target recognition on the target image to obtain location information of the target object; The UAV adjusts the flight attitude of the UAV according to the position information of the target object; The drone sends the target image to the ground station.

6. The method according to claim 5, characterized in that The drone performs target recognition on the target image to obtain location information of the target object, including: Convert the color image to obtain a grayscale image; Performing Gaussian filtering and smoothing processing on the grayscale image to obtain a smoothed grayscale image; Performing edge detection on the smoothed grayscale image to obtain edge information of the image; Post-processing and Hough transform algorithm are used to extract edge information and determine the location information of the target object.

7. The method according to claim 5, characterized in that The method further comprises: The drone identifies the charging area marker in the charging area; By calculating the relative position between the gimbal and the center point of the charging area marker, the deviation between the current position of the drone and the charging area is obtained; The flight attitude and position of the drone are adjusted according to the deviation between the current position of the drone and the charging area to obtain the fixed landing position.

8. The method according to claim 5, characterized in that The method further comprises: The target image transmitted by the UAV is used to train the landmark recognition model to obtain the landmark recognition model; Using the landmark recognition model to perform landmark recognition on the newly input image to obtain a recognition result; The recognition results are sorted to obtain a recognition result statistics table.

Images