Unmanned aerial vehicle laser radar remote sensing equipment for cotton field and data acquisition method

Abstract

The invention discloses an unmanned aerial vehicle laser radar remote sensing device for a cotton field and a data acquisition method.The device comprises an unmanned aerial vehicle, a laser radar, a loading mechanism, a radar micro-host and a ground control module, and the loading mechanism comprises a supporting column, a lower supporting plate and an upper supporting plate; the upper end of the supporting column is connected with the unmanned aerial vehicle through the connecting ring, the lower end of the supporting column is connected with the lower supporting plate through the base, the upper supporting plate is installed in the middle of the lower supporting plate, the laser radar is installed on the radar sleeve, and the radar sleeve is installed on the upper supporting plate through the holder. The laser radar is connected with the radar micro host; the ground control module is connected with the radar micro host. According to the method, more comprehensive and abundant cotton phenotype information can be provided, the method is of great significance to scientific management such as cotton growth monitoring, yield estimation and precise pesticide spraying, invasion to crops is reduced, and flexible collection is achieved.

Description

Technical Field

[0001] The invention relates to unmanned aerial vehicle technology, and in particular to an unmanned aerial vehicle laser radar remote sensing device for a cotton field and a data collection method. Background Art

[0002] Cotton is one of the most important crops in the world and an important source of natural fiber for the global textile industry. It has important uses in defense, medicine, and the automotive industry, and has no economic substitute. Upland cotton is the main variety of cotton, accounting for more than 90% of the world's annual production and has a high economic value. Crop height, canopy porosity, and canopy volume are important plant phenotypic structural parameters that are closely related to crop growth, health, yield estimation, and biomass estimation. Therefore, it is of great significance to obtain the phenotypic structural parameters of large-scale cotton.

[0003] Various 2D imaging techniques have been used to extract canopy parameters non-invasively by segmenting image pixels into plant components (such as leaves and stems). However, some features (such as plant height and canopy volume) may not be extracted from available 2D images due to vegetation self-occlusion, leaf crossing, and lack of depth information. Moreover, optical images cannot directly provide forest vertical structural parameters and may be affected by saturation effects when used alone. With the recent development of LiDAR technology, light detection and ranging (LiDAR) is currently considered to be the most accurate technology for detecting canopy structure, and 3D point cloud data is gradually being used in various agricultural scenarios. LiDAR actively emits laser pulses and relies on the penetration of pulses to detect the three-dimensional structural information of vegetation. The laser beam emitted by the LiDAR sensor has the advantages of high energy density, small divergence angle, and long linear propagation distance. Compared with RGB cameras and depth cameras, it is not affected by light factors such as strong light and shadows, and has rich geometric information. It can penetrate vegetation to obtain vegetation position and internal structure, and more accurately obtain phenotypic information such as vegetation height, crown width, and canopy volume.

[0004] Traditionally, cotton phenotypic information is obtained through manual field measurements, but field measurement methods are time-consuming, labor-intensive, destructive, and invasive, and are only applicable to small areas, so it is difficult to quickly obtain cotton canopy information over a large area. Vehicle-mounted sensor equipment has a single perspective that is greatly affected by the environment and has poor mobility. If the road in the field is rugged, it will be affected by the terrain, thereby reducing performance. Summary of the invention

[0005] The first purpose of the present invention is to overcome the deficiencies of the above prior art and provide a UAV laser radar remote sensing device for cotton fields. The UAV laser radar remote sensing device for cotton fields improves scanning efficiency, reduces invasion of crops, and detects data accurately.

[0006] The second object of the present invention is to provide a method for collecting cotton field location cloud data based on drones.

[0007] The objective of the present invention is achieved through the following technical scheme: The unmanned aerial vehicle laser radar remote sensing equipment for cotton fields comprises an unmanned aerial vehicle, a laser radar, a loading mechanism, a radar microhost and a ground control module, wherein the loading mechanism comprises a support column, a lower support plate and an upper support plate, the upper end of the support column is connected to the unmanned aerial vehicle through a connecting ring, the lower end of the support column is connected to the lower support plate through a base, the upper support plate is installed in the middle of the lower support plate, the laser radar is installed on a radar sleeve, and the radar sleeve is installed on the upper support plate through a pan-tilt platform; the laser radar is connected to the radar microhost; the ground control module is connected to the radar microhost.

[0008] Preferably, the upper support plate is provided with a pair of arc-shaped protrusions for mounting the pan / tilt head, and the arc angle of the arc-shaped protrusions is θ, 45°≤θ≤75°.

[0009] Preferably, the base is provided with a pair of outer slot holes and a pair of inner slot holes, the two outer slot holes are symmetrically arranged relative to the central axis of the base, and the two inner slot holes are symmetrically arranged relative to the central axis of the base; the outer slot holes and the inner slot holes are both arc-shaped, wherein the arc angle of the outer slot holes is α, 45°≤α≤65°, and the arc angle of the inner slot holes is β, 45°≤β≤65°.

[0010] Preferably, the radar sleeve is provided with an inner cavity matching the shape of the laser radar, two opposite side walls of the radar sleeve are provided with a first mounting hole for fixing the laser radar, and the third side wall of the radar sleeve is provided with a second mounting hole for fixing the gimbal.

[0011] Preferably, the connecting ring includes a lower half ring and an upper half ring, the lower half ring is fixed to the upper end of the support column, the upper half ring is connected to the lower half ring through a screw, and the upper half ring and the lower half ring clamp the arm of the drone.

[0012] Preferably, both ends of the lower support plate are provided with strip-shaped slot holes for mounting the base.

[0013] The method for collecting location cloud data of cotton fields based on drones adopts the drone laser radar remote sensing device for cotton fields described in the first invention object, and comprises the following steps:

[0014] S1. According to the disturbance of the UAV rotor flow field on the cotton and the laser radar detection angle, combined with the two-dimensional terrain of the area to be measured, the measurement points and flight altitude are set, and the route is planned;

[0015] S2, using a drone equipped with a laser radar to reach each measurement point in sequence from the measurement point at the first point along the route;

[0016] S3, the laser radar works during the route and generates a global point cloud data package, performs image processing on the point cloud data package, and generates a three-dimensional cotton point cloud global map;

[0017] S4. CloudCompare point cloud processing software was used to segment the global map of cotton point cloud and extract the data of individual cotton plants. The plant height, porosity and canopy volume of individual cotton plants were calculated using the Python algorithm library and Matlab data processing software.

[0018] Preferably, step S3 comprises the following steps:

[0019] S31. Use the ROS robot operating system in combination with the radar microhost to collect cotton point cloud data, and record the data points collected by each frame of the laser radar to form a point cloud data packet;

[0020] S32, obtain IMU and lidar pose information through SLAM mapping algorithm Fast-lio2 to correct the point cloud data packet;

[0021] S33. Utilize a statistical outlier removal algorithm to remove redundant point clouds from the point cloud data to generate a three-dimensional cotton point cloud global map.

[0022] Preferably, there are four measuring points, which are respectively located at the four corners of the cotton field. The four measuring points are connected in sequence in a counterclockwise direction from a bird's-eye view to form a route, wherein the first point is the starting point and the fourth point is the end point.

[0023] Preferably, the UAV flying altitude is 10-15m and the travel speed is 2-3m / s.

[0024] The present invention has the following advantages over the prior art:

[0025] 1. The present invention can provide high-precision geographic information of cotton fields, including topography, landform, canopy information, etc., through the combination of drones and laser radar technology. It can provide more comprehensive and rich cotton phenotypic information, which is of great significance to scientific management such as cotton growth monitoring, yield estimation, and precise spraying of pesticides.

[0026] 2. The present invention combines drone and lidar technology to quickly acquire cotton phenotypic data. Compared with traditional manual data collection and vehicle-mounted equipment collection, it reduces the invasion of crops, has richer perspectives, more flexible collection routes, richer scanning data, higher collection efficiency, and saves investment. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 It is a structural schematic diagram of the UAV laser radar remote sensing device for cotton fields of the present invention.

[0028] Figure 2 It is a structural schematic diagram of the loading mechanism of the present invention.

[0029] Figure 3 It is a structural schematic diagram of the upper support plate of the present invention.

[0030] Figure 4 It is a schematic diagram of a base of the present invention.

[0031] Figure 5 It is a structural schematic diagram of the radar sleeve of the present invention.

[0032] Figure 6 It is a flow chart of the cotton field location cloud data collection method based on unmanned aerial vehicle of the present invention.

[0033] Figure 7 It is a schematic diagram of the measurement points and routes of the drone of the present invention.

[0034] Figure 8 It is the cotton point cloud data collected and generated by the present invention.

[0035] Fig. 9 It is a schematic diagram of partial cotton individual plant point cloud extraction of the present invention.

[0036] Among them, 1 is a UAV, 2 is a laser radar, 3 is a loading mechanism, 4 is a radar microcomputer, 5 is a ground control module, 6 is a support column, 7 is an upper support plate, 8 is a lower support plate, 9 is a connecting ring, 10 is a base, 11 is a radar sleeve, 12 is a gimbal, 13 is a radar power supply, 14 is a UAV battery, 15 is an arc-shaped convex body, 16 is an outer slot hole, 17 is an inner slot hole, 18 is a first mounting hole, 19 is a second mounting hole, 20 is a lower half ring, 21 is an upper half ring, and 22 is an open slot. DETAILED DESCRIPTION

[0037] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0038] like Figures 1 to 4 As shown, an unmanned aerial vehicle (UAV) laser radar remote sensing device for a cotton field includes a UAV, a laser radar, a loading mechanism, a radar micro-host and a ground control module. The loading mechanism includes a support column, a lower support plate and an upper support plate. The upper end of the support column is connected to the UAV through a connecting ring, and the lower end of the support column is connected to the lower support plate through a base. The upper support plate is installed in the middle of the lower support plate. The laser radar is installed in a radar sleeve, and the radar sleeve is installed on the upper support plate through a gimbal; the laser radar is connected to the radar micro-host; and the ground control module is connected to the radar micro-host.

[0039] Specifically, two support columns are used, and the two ends of the lower support plate are connected to the two arms of the drone through the two support columns, so as to ensure the stability of the laser radar installed on the upper support plate through the base and the gimbal. In order to facilitate the control of the flight of the drone, a remote controller and a drone battery that provides energy for the drone are also included. The operator controls the flight trajectory of the drone through the remote controller according to the determined route to ensure accurate data collection. The ground control module in this embodiment uses a computer.

[0040] At the same time, there is a radar power supply that provides power for the radar micro-host and the laser radar. This radar power supply and the radar micro-host are packaged together and then installed on one side of the drone. The radar power supply, radar micro-host and laser radar are connected by lines. For the flight of the drone, a drone battery is set on the top of the drone to give the drone a good endurance effect. During the work process, the ROS robot control system in the radar micro-host controls the laser radar in real time to complete the recording of the cotton point cloud data packet. One of the operators uses a remote control to control the flight altitude, flight route and flight speed of the drone. The other operator refers to the real-time position of the drone and uses the remote operation software to remotely control the micro-host through the ground control module to ensure the accuracy and continuity of the data packet recording. During the flight of the drone, the radar micro-host is remotely controlled by the ground control computer to monitor the execution of the task and control it in real time to realize the collection of cotton field data packets.

[0041] The upper support plate is provided with a pair of arc-shaped protrusions for installing the pan-tilt head. The arc angle of the arc-shaped protrusions is θ, 45°≤θ≤75°. While fixing the position of the mechanical pan-tilt head, it is convenient for loading and unloading the pan-tilt head and can also save materials.

[0042] The base is provided with a pair of outer slots and a pair of inner slots, the two outer slots are symmetrically arranged relative to the central axis of the base, and the two inner slots are symmetrically arranged relative to the central axis of the base; the outer slots and the inner slots are both arc-shaped, wherein the arc angle of the outer slots is α, 45°≤α≤65°, and the arc angle of the inner slots is β, 45°≤β≤65°. The angles between the arms of different UAVs (four-axis, six-axis and eight-axis UAVs, etc.) will be different. The connection of the support column to the arms of UAVs with different numbers of axes will affect the connection position of the support column base hole position and the support plate. Figure 3 As shown, the support column can be rotated in the vertical direction to adjust the angle. No matter what the angle between the two arms is, it can ensure that there are corresponding holes between the support column base and the support plate for bolt connection. It is suitable for various types of drones and has strong versatility.

[0043] like Figure 5As shown, the radar sleeve is provided with an inner cavity matching the shape of the laser radar, and the two opposite side walls of the radar sleeve are provided with a first mounting hole for fixing the laser radar, and the third side wall of the radar sleeve is provided with a second mounting hole fixed to the pan-tilt. This structure is simple and ensures the stability of the laser radar after installation to determine the accuracy of data acquisition. Among them, there are 4 first mounting holes set on each side wall, and the diameter of these four first mounting holes is 5mm, which are distributed in a matrix. The distances between the transverse hole spacing and the longitudinal hole spacing are 45mm and 64mm respectively, so as to ensure that the laser radar can be firmly fixed in the inner cavity of the radar sleeve. There are 2 second mounting holes with an aperture of 8mm and a spacing of 20mm between the two holes, so as to facilitate the rapid fixing of the radar sleeve to the pan-tilt. At the same time, in order to facilitate wiring, an open slot is set on one of the side walls of the radar sleeve. This open slot is square with a side length of 28 to 32mm.

[0044] The connecting ring includes a lower half ring and an upper half ring, the lower half ring is fixed to the upper end of the support column, the upper half ring is connected to the lower half ring by a screw, and the upper half ring and the lower half ring clamp the arm of the drone. The lower half ring and the upper half ring of the connecting ring clamp the arm of the drone from the top and bottom respectively to ensure the stability of the connection. At the same time, both ends of the lower support plate are provided with strip slots for mounting the base. This facilitates the spacing between the two support columns to be adjusted according to the arm of the drone to meet the needs of drones of different sizes.

[0045] like Figure 6 As shown, the method for collecting location cloud data of cotton fields based on drones, using the drone laser radar remote sensing device for cotton fields described in the first invention object, includes the following steps:

[0046] S1. According to the disturbance of the UAV rotor flow field on the cotton and the laser radar detection angle, combined with the two-dimensional terrain of the area to be measured, the measurement points and flight altitude are set, and the route is planned;

[0047] Specifically, after the drone has built the drone-based laser radar remote sensing equipment, the route is planned. Figure 7 As shown, during the laser radar scanning process carried by the drone, the drone's flight altitude is 12m, the travel speed is 2m / s, and the flight path is U-shaped.

[0048] S2. Use a drone equipped with a laser radar to reach each measuring point in sequence from the measuring point located at the first point along the route; there are four measuring points, and these four measuring points are located at the four corners of the cotton field respectively. The four measuring points are connected in a counterclockwise direction from a bird's-eye view to form a route, where the first point is the starting point and the fourth point is the end point. In the process of the drone flying to each measuring point in sequence along the route, the laser radar collects cotton point cloud data, and sets the distance resolution to 0.02m and the angle resolution to 0.05°. Before the laser radar scans, avia is preheated for a few minutes, and the sampling frequency is set to 240khz. The vertical scanning angle of the laser radar is adjusted between 30° and 60° through the radar two-axis mechanical gimbal, which can capture cotton canopy information from multiple angles while reducing the problem of cotton self-occlusion.

[0049] S3, the laser radar works during the route and generates a global point cloud data package, performs image processing on the point cloud data package, and generates a three-dimensional cotton point cloud global map;

[0050] Step S3 includes the following steps:

[0051] S31. Use the ROS robot operating system in combination with the radar microhost to collect cotton point cloud data, and record the data points collected by each frame of the laser radar to form a point cloud data packet;

[0052] S32, obtain IMU and lidar pose information through SLAM mapping algorithm Fast-lio2 to correct the point cloud data packet;

[0053] S33, using a statistical outlier removal algorithm to remove redundant point clouds from the point cloud data to generate a three-dimensional cotton point cloud global map, such as Figure 8 shown.

[0054] Specifically, experimenter 1 used a remote controller to control the flight altitude, flight route, and flight speed of the drone. Experimenter 2 referred to the real-time position of the drone and controlled the micro-host remotely using remote operation software through a ground control computer to ensure the accuracy and continuity of data packet recording.

[0055] S4. Use CloudCompare point cloud processing software to segment the global map of cotton point cloud and extract single cotton plant data. Some of the extracted single cotton plant data are as follows: Fig. 9 Then, the height, porosity and canopy volume of a single cotton plant are calculated through the Python algorithm library and Matlab data processing software, thus providing data support for cotton growth monitoring.

[0056] The drone's flight altitude is 10-15m and its travel speed is 2-3m / .

[0057] After adopting the above scheme, compared with the existing manual measurement technology and vehicle-mounted collection method, the present invention solves the problems of traditional data collection equipment such as single viewing angle, poor maneuverability, strong invasiveness, complex structure and high cost due to the combination of drone and lidar, thereby achieving the effect of improving scanning efficiency, enriching data collection results, reducing invasion of crops and saving investment.

[0058] The above specific implementation modes are preferred embodiments of the present invention and cannot be used to limit the present invention. Any other changes or other equivalent replacement methods that do not deviate from the technical solution of the present invention are included in the protection scope of the present invention.

Claims

1. UAV laser radar remote sensing equipment for cotton fields, characterized in that: It includes an unmanned aerial vehicle, a laser radar, a loading mechanism, a radar micro-host and a ground control module. The loading mechanism includes a support column, a lower support plate and an upper support plate. The upper end of the support column is connected to the unmanned aerial vehicle through a connecting ring, and the lower end of the support column is connected to the lower support plate through a base. The upper support plate is installed in the middle of the lower support plate. The laser radar is installed on the radar sleeve, and the radar sleeve is installed on the upper support plate through a gimbal; the laser radar is connected to the radar micro-host; and the ground control module is connected to the radar micro-host.

2. The unmanned aerial vehicle laser radar remote sensing device for cotton fields according to claim 1 is characterized in that: The upper support plate is provided with a pair of arc-shaped convex bodies for mounting the pan / tilt head, and the arc angle of the arc-shaped convex bodies is θ, 45°≤θ≤75°.

3. The UAV laser radar remote sensing device for cotton fields according to claim 1 is characterized in that: The base is provided with a pair of outer slot holes and a pair of inner slot holes, the two outer slot holes are symmetrically arranged relative to the central axis of the base, and the two inner slot holes are symmetrically arranged relative to the central axis of the base; the outer slot holes and the inner slot holes are both arc-shaped, wherein the arc angle of the outer slot holes is α, 45°≤α≤65°, and the arc angle of the inner slot holes is β, 45°≤β≤65°.

4. The UAV laser radar remote sensing device for cotton fields according to claim 1 is characterized in that: The radar sleeve is provided with an inner cavity matching the shape of the laser radar, two opposite side walls of the radar sleeve are provided with a first mounting hole for fixing the laser radar, and the third side wall of the radar sleeve is provided with a second mounting hole fixed to the gimbal.

5. The UAV laser radar remote sensing device for cotton fields according to claim 1, characterized in that: The connecting ring comprises a lower half ring and an upper half ring, wherein the lower half ring is fixed to the upper end of the supporting column, the upper half ring is connected to the lower half ring through a screw rod, and the upper half ring and the lower half ring clamp the arm of the drone.

6. The unmanned aerial vehicle laser radar remote sensing device for cotton fields according to claim 1, characterized in that: Both ends of the lower support plate are provided with strip-shaped slot holes for installing the base.

7. A method for collecting cotton field location cloud data based on drones, characterized in that: The unmanned aerial vehicle laser radar remote sensing device for cotton fields according to any one of claims 1 to 6 comprises the following steps: S1. According to the disturbance of the UAV rotor flow field on the cotton and the laser radar detection angle, combined with the two-dimensional terrain of the area to be measured, the measurement points and flight altitude are set, and the route is planned; S2, using a drone equipped with a laser radar to reach each measurement point in sequence from the measurement point at the first point along the route; S3, the laser radar works during the route and generates a global point cloud data package, performs image processing on the point cloud data package, and generates a three-dimensional cotton point cloud global map; S4. CloudCompare point cloud processing software was used to segment the global map of cotton point cloud and extract the data of individual cotton plants. The plant height, porosity and canopy volume of individual cotton plants were calculated using the Python algorithm library and Matlab data processing software.

8. The method for collecting cotton field location cloud data based on drone according to claim 7, characterized in that: Step S3 includes the following steps: S31. Use the ROS robot operating system in combination with the radar microhost to collect cotton point cloud data, and record the data points collected by each frame of the laser radar to form a point cloud data packet; S32, obtain IMU and lidar pose information through SLAM mapping algorithm Fast-lio2 to correct the point cloud data packet; S33. Utilize a statistical outlier removal algorithm to remove redundant point clouds from the point cloud data to generate a three-dimensional cotton point cloud global map.

9. The method for collecting cotton field location cloud data based on drone according to claim 7, characterized in that: There are four measuring points, which are respectively located at the four corners of the cotton field. The four measuring points are connected in a counterclockwise direction from a bird's-eye view to form a route, wherein the first point is the starting point and the fourth point is the end point.

10. The method for collecting cotton field location cloud data based on drone according to claim 7, characterized in that: The drone's flight altitude is 10-15m and its travel speed is 2-3m / s.

Images