Soil sampling device and method for detecting organic carbon in soil

Abstract

The invention belongs to the technical field of soil detection, and discloses a soil sampling device and method for soil organic carbon detection, and the device comprises an unmanned aerial vehicle main body which is provided with a caster assembly; a storage seat is fixedly arranged below the unmanned aerial vehicle main body; a plurality of sampling barrels are arranged in the storage seat in a sliding manner; a driving mechanism is fixedly arranged on the unmanned aerial vehicle main body; a connecting rod is rotationally arranged on the driving mechanism; a transmission mechanism is rotationally arranged on the unmanned aerial vehicle main body; the connecting rod is in transmission connection with the transmission mechanism; a transposition mechanism is fixedly arranged on the storage seat; a pushing mechanism is rotatably arranged on the storage seat; the pushing mechanism is in transmission connection with the transposition mechanism; a plurality of rubber plugs are arranged on the pushing mechanism; a monitoring mechanism is arranged on the unmanned aerial vehicle body. The sampling barrel is automatically replaced and sealed through the transposition mechanism, underground foreign matter interference is avoided, equipment is protected, and invalid operation is reduced; the whole process does not need manual operation; and the image analysis module identifies sampling abnormity in real time.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of soil detection, in particular to a soil sampling device and method for soil organic carbon detection. BACKGROUND

[0002] The soil sampling device for soil organic carbon detection is a tool or equipment specially designed for collecting soil samples for organic carbon content detection. The main function of the device is to collect soil samples with representativeness and accuracy for subsequent accurate detection of organic carbon content. Through the collected soil samples, the organic carbon content in the soil can be analyzed, and the quality and fertility of the soil can be evaluated, thereby providing important data support for agricultural management, environmental monitoring and soil scientific research.

[0003] The document with the prior art publication number CN119334690A provides a soil sampling device for soil organic carbon detection, which comprises a protective shell. A hydraulic telescopic rod is fixedly installed at the middle end inside the protective shell. The cooperation of the foot pedal mechanism and the linkage enables the linkage to link the foot pedal mechanism and the moving mechanism, thereby ensuring the portable movement and stable fixation of the device. An operator steps on the foot pedal, and the sliding plate connected by the spring slides on the protective shell. The sliding block at the bottom surface of the sliding plate slides in the sliding groove on the outer surface of the protective shell, ensuring stable movement of the sliding plate. The U-shaped plate fixedly connected to the top surface of the sliding plate drives the connecting rod to move, and the connecting rod drives the rotating rod to rotate around the rotating shaft. Since the two ends of the rotating shaft are connected to one side of the guide rail through the support, the lifting rod movably connected to one end of the rotating rod slides upward in the guide rail, and the connecting column at the bottom surface of the lifting rod drives the base and the universal wheel fixedly connected to the bottom surface to rise.

[0004] Although the prior art solution in the above can achieve the beneficial effects related to the prior art, it still has the following defects: 1. The prior art lacks effective sealing design during soil organic carbon sampling, and the sample is easy to contact with the outside world, which leads to easy volatilization or contamination of organic carbon, and it is difficult to guarantee the original state of the sample before detection. 2. The replacement and sealing of the sampling cylinder in the prior art require manual intervention, which leads to a long sampling period for a single group of samples, and cannot meet the batch sampling demand of a large area. 3. The prior art lacks the functions of pre-detection of underground foreign objects and dynamic adjustment of the path, and the equipment is easy to collide with underground foreign objects and be damaged during sampling, resulting in high equipment failure rate and a lot of invalid work.

[0005] In view of this, we propose a soil sampling device and method for soil organic carbon detection. SUMMARY

[0006] 1. Technical problem to be solved

[0007] The purpose of the present application is to provide a soil sampling device and method for soil organic carbon detection, which solves the technical problems raised in the background art, realizes automatic replacement and sealing of the sampling cylinder by the transposition mechanism, without manual intervention, shortens the sampling cycle of a single sample, avoids interference from underground foreign objects, protects the equipment and reduces invalid operations, avoids pollution caused by hand contact with the sample, and reduces random errors caused by human operation; the image analysis module can identify sampling abnormalities in real time and respond in a timely manner.

[0008] 2. Technical solution

[0009] The technical solution of the present application provides a soil sampling device for soil organic carbon detection, which comprises a UAV main body, a caster assembly, a storage seat, a sampling cylinder, a driving mechanism, a transmission mechanism, a connecting rod, a transposition mechanism, a pushing mechanism, a clamping seat and a monitoring mechanism.

[0010] The UAV main body is provided with a caster assembly; the UAV main body is fixedly provided below with a storage seat; a plurality of sampling cylinders are slidably arranged in the storage seat; the UAV main body is fixedly provided with a driving mechanism; the driving mechanism is rotatably provided with a connecting rod; the UAV main body is rotatably provided with a transmission mechanism; the connecting rod is in transmission connection with the transmission mechanism; and the connecting rod is in detachable fixed connection with the sampling cylinder.

[0011] The storage seat is fixedly provided with a transposition mechanism; the storage seat is rotatably provided with a pushing mechanism; the pushing mechanism is in transmission connection with the transposition mechanism; and the pushing mechanism is provided with a plurality of rubber plugs.

[0012] The storage seat is fixedly provided with a clamping seat; the clamping seat is in transmission connection with the transposition mechanism. The UAV main body is fixedly provided with a monitoring mechanism, which monitors the soil sampling process and discovers abnormal conditions in a timely manner.

[0013] Through the above technical solution, the UAV main body flies to the position where sampling is needed, the driving mechanism drives the sampling cylinder to move downward, at the same time, the transmission mechanism drives the sampling cylinder to rotate, so that the sampling cylinder extends into the underground, and the soil sampling operation is completed. Then, the driving mechanism drives the sampling cylinder to move upward to a predetermined position; the transposition mechanism drives the pushing mechanism to rotate, so that one of the rubber plugs is below the sampling cylinder; the driving mechanism drives the sampling cylinder to move downward, so that the rubber plug is clamped into the bottom of the sampling cylinder for sealing operation; then the driving mechanism drives the sampling cylinder to move upward by a distance, and the connecting rod cancels the connection with the sampling cylinder. The transposition mechanism drives the pushing mechanism to rotate by one hundred and eighty degrees, and drives the clamping to rotate by a certain angle, so that the sampling cylinder containing the soil sample moves away from the position below the connecting rod, and the next sampling cylinder rotates to the position below the connecting rod for standby. The monitoring mechanism monitors the soil sampling process and discovers abnormal conditions in a timely manner.

[0014] As an optional solution of the present application, the caster assembly comprises an electric telescopic rod, a connecting seat, a sleeve and a caster;

[0015] The four machine legs are fixedly arranged below the unmanned aerial vehicle body; the electric telescopic rods are fixedly arranged between the two machine legs on the front side and the two machine legs on the rear side; the connecting seats are fixedly arranged on the movable rods of the electric telescopic rods; the sleeves are fixedly arranged on the two sides of the connecting seats; the sleeves are slidably arranged on the machine legs; the casters are rotatably arranged on the two sides of the connecting seats. The casters are driven by electricity. The laser range finder is fixedly arranged below the connecting seat.

[0016] As an optional solution of the present application, the driving mechanism comprises a motor A, a bevel gear A, a spur gear A and a toothed plate;

[0017] The motor A is fixedly arranged on the unmanned aerial vehicle body; the coaxial bevel gear A and the coaxial spur gear A are fixedly arranged on the output end of the motor A; the guide pipe is fixedly arranged on the unmanned aerial vehicle body; the toothed plate is slidably arranged in the guide pipe; the toothed plate is in meshing transmission connection with the spur gear A. The connecting rod is rotatably arranged on the toothed plate through the bearing; the bevel gear A is in transmission connection with the transmission mechanism.

[0018] As an optional solution of the present application, the transmission mechanism comprises a toothed column and a bevel gear B;

[0019] The toothed column is rotatably arranged on the unmanned aerial vehicle body; the coaxial bevel gear B is arranged on the upper end of the toothed column; the bevel gear B is in meshing transmission connection with the bevel gear A.

[0020] The toothed ring A is fixedly arranged on the upper end of the connecting rod; the toothed ring A is in meshing transmission connection with the toothed column.

[0021] As an optional solution of the present application, the transposition mechanism comprises a rotating seat, a spur gear B, a spur gear C and a motor B; the rotating seat is rotatably arranged on the storage seat; the inner tooth groove is formed in the inner side of the rotating seat; the two groups of driving teeth A are symmetrically arranged on the lower part of the outer side of the rotating seat; one group of driving teeth B is arranged above one group of driving teeth A;

[0022] The motor B is fixedly arranged on the storage seat; the coaxial spur gear B is fixedly arranged on the output end of the motor B; the coaxial spur gear C is rotatably arranged on the storage seat; the spur gear C is in meshing transmission connection with the clamping seat; the plurality of tooth grooves are arranged in the inner side of the clamping seat; the spur gear B is in meshing transmission connection with the inner tooth groove; the spur gear C is in meshing transmission connection with the driving teeth B. The driving teeth A are in transmission connection with the pushing mechanism. The plurality of arc clamping grooves are uniformly arranged on the outer side of the clamping seat; the arc clamping grooves are matched with the sampling cylinder.

[0023] As an optional solution of the present application, the pushing mechanism comprises a cylinder, a toothed ring B, a positioning pipe, a supporting seat, a rubber plug and an electric push rod;

[0024] The storage seat is fixedly provided with a positioning pipe, and a cylinder is rotatably arranged on the positioning pipe; a plurality of rubber plugs are arranged in the cylinder, a top cover is detachably arranged on the top of the cylinder, and a spring is fixedly arranged below the top cover. The spring is in abutment with the uppermost rubber plug. A tooth ring B is detachably and fixedly arranged on the cylinder; the tooth ring B is in meshing transmission connection with the driving teeth A.

[0025] A supporting seat is fixedly arranged at the lower end of the cylinder; an electric push rod is arranged on the supporting seat, and a push plate is fixedly arranged on the movable rod of the electric push rod; the push plate is in abutment with the rubber plug. A square hole is formed in the upper layer of one end of the supporting seat.

[0026] As an optional solution of the application, the top of the sampling cylinder is provided with a square groove; the lower end of the connecting rod is fixedly provided with a square column, and an electromagnet is fixedly arranged in the square column; the size and shape of the square column are matched with the square groove; the square column can be inserted into the square groove, and the sampling cylinder and the connecting rod are fixedly connected through the electromagnet. The square groove is provided with a ferromagnetic gasket at the bottom, and the electromagnet of the connecting rod forms a magnetic attraction fit.

[0027] The top of the sampling cylinder is provided with a ring groove, and a plurality of air guide holes are arranged in the ring groove; a nitrile rubber sealing ring is arranged in the ring groove; when the square column is inserted into the square groove, the sealing ring is deformed by being extruded by the annular groove at the bottom of the connecting rod, so that the air path is sealed. The other end of the sampling cylinder is provided with a plurality of air bag storage grooves, and an air bag is fixedly arranged in each air bag storage groove; the air guide holes are in communication with the air bags. In the initial state, the air bag is attached to the air bag storage groove, and the outer side is aligned with the inner wall of the sampling cylinder. The bottom of the connecting rod is provided with an annular groove; a gas pump is fixedly arranged on the main body of the unmanned aerial vehicle, and the output end of the gas pump is in communication with the annular groove at the bottom of the connecting rod. The output end of the gas pump is fixedly provided with a gas pressure gauge. A plurality of sawteeth are arranged below the connecting rod.

[0028] Further, the monitoring mechanism comprises:

[0029] The data collection module collects geological data and vegetation data of the soil sampling area; historical data is collected, and the data is labeled.

[0030] The route planning module reasonably plans the data and position of the sampling points according to the geological data and sampling requirements of the soil sampling area, and plans the optimal flight and walking route of the unmanned aerial vehicle main body;

[0031] The image acquisition module includes a plurality of high-definition cameras for acquiring high-definition images of the soil sampling process;

[0032] The image analysis module analyzes and identifies the collected images, and identifies abnormal conditions in the soil sampling process in a timely manner;

[0033] The ground penetrating radar is fixedly arranged on the main body of the unmanned aerial vehicle, emits high-frequency electromagnetic waves, identifies foreign matters through the electromagnetic characteristic difference of different media, and can generate underground profile images. Foreign matters such as stones, roots and pipelines are identified.

[0034] The weather data collection module: collect the weather data of the soil sampling area in real time;

[0035] The radar data analysis module: analyze the radar data and timely identify the foreign matter under the soil sampling area.

[0036] The alarm module: including an alarm, when monitoring abnormal conditions, timely alarm.

[0037] The control center: with the data collection module, route planning module, image acquisition module, image analysis module, ground penetrating radar, radar data analysis module and alarm module.

[0038] The application provides a soil sampling method for soil organic carbon detection, comprising the following steps:

[0039] S1, the data collection module of the monitoring mechanism collects the geological data and vegetation data of the soil sampling area; collect historical data and label the data.

[0040] S2, the route planning module reasonably plans the data and position of the sampling point according to the geological data of the soil sampling area and the sampling requirement, and plans the optimal unmanned aerial vehicle main body flight and walking route;

[0041] S3, the unmanned aerial vehicle main body carries the ground penetrating radar to scan the sampling area. The radar data analysis module analyzes the radar data and timely identifies the foreign matter under the soil sampling area. The weather data collection module collects the weather data of the soil sampling area in real time;

[0042] S4, the route planning module adjusts the flight and walking route according to the analysis result of the radar data analysis module;

[0043] S5, the unmanned aerial vehicle main body makes the sampling cylinder sample the soil through the driving mechanism, transmission mechanism, connecting rod, transposition mechanism, pushing mechanism and clamping seat.

[0044] S6, the image acquisition module collects high-definition images during the soil sampling process;

[0045] S7, the image analysis module analyzes and identifies the collected images, and timely identifies the abnormal conditions during the soil sampling process;

[0046] S8, when monitoring abnormal conditions, the alarm module timely alarms.

[0047] 3. Beneficial effects

[0048] One or more technical solutions provided in the technical scheme of the application have at least the following technical effects or advantages:

[0049] 1. The present application is aimed at the characteristics of soil organic carbon being easily disturbed. In the sampling process, the rubber plug is used to seal the inert material sampling cylinder, which reduces the contact between the sample and the outside world, avoids the volatilization or pollution of organic carbon, and protects the original state of the sample before detection.

[0050] 2. Improve sampling efficiency and reduce operation cost: the route planning module combines the mixed mode of unmanned aerial vehicle flight and wheel walking, quickly reaches the destination by long-distance flight, and saves energy by walking in short distance. Compared with traditional manual sampling or pure flight mode, the operation efficiency is improved, and more sampling points can be completed in a single operation.

[0051] 3. The transposition mechanism realizes automatic replacement and sealing of the sampling cylinder without manual intervention, shortens the sampling period of a single group of samples, and is suitable for batch sampling demand in large area.

[0052] 4. Avoid interference from underground foreign objects, protect equipment and reduce invalid operation: the ground penetrating radar and radar data analysis module scans underground foreign objects in advance, and the route planning module dynamically adjusts the path to avoid equipment collision and damage during sampling, and reduces the equipment failure rate.

[0053] 5. Realize full-process automation and intelligentization, reduce human error: from sampling point planning, foreign object detection, automatic sampling to anomaly monitoring, the whole process does not need manual operation, avoids pollution caused by hand contact with the sample, and reduces the randomness error of human operation. The image analysis module identifies sampling abnormalities in real time and responds in time to ensure that problems are quickly handled and the effective sample rate is improved. BRIEF DESCRIPTION OF DRAWINGS

[0054] Figure 1 The overall schematic diagram of the soil sampling device for soil organic carbon detection disclosed in a preferred embodiment of the present application;

[0055] Figure 2 The driving mechanism and transmission mechanism schematic diagram of the soil sampling device for soil organic carbon detection disclosed in a preferred embodiment of the present application;

[0056] Figure 3 The overall schematic diagram of the pushing mechanism of the soil sampling device for soil organic carbon detection disclosed in a preferred embodiment of the present application;

[0057] Figure 4 The pushing mechanism schematic diagram of the soil sampling device for soil organic carbon detection disclosed in a preferred embodiment of the present application;

[0058] Figure 5 The transfer seat schematic diagram of the soil sampling device for soil organic carbon detection disclosed in a preferred embodiment of the present application;

[0059] Figure 6 The sampling cylinder schematic diagram of the soil sampling device for soil organic carbon detection disclosed in a preferred embodiment of the present application;

[0060] Reference signs: 1, unmanned aerial vehicle body; 2, trundle assembly; 3, storage seat; 4, sampling cylinder; 5, driving mechanism; 6, transmission mechanism; 7, connecting rod; 8, transposition mechanism; 9, pushing mechanism; 10, clamping seat; 11, guide pipe; 12, leg; 21, electric telescopic rod; 22, connecting seat; 23, sleeve; 24, trundle; 31, round hole; 41, square groove; 42, ring groove; 43, air guide hole; 44, air bag storage groove; 51, motor A; 52, bevel gear A; 53, spur gear A; 54, toothed plate; 61, toothed column; 62, bevel gear B; 71, toothed ring A; 81, rotating seat; 811, inner tooth groove; 812, driving tooth A; 813, driving tooth B; 82, spur gear B; 83, spur gear C; 84, motor B; 91, cylinder; 92, toothed ring B; 93, positioning pipe; 94, supporting seat; 95, rubber plug; 96, square hole; 97, electric push rod; 101, arc clamping groove. DETAILED DESCRIPTION

[0061] The application will be further described in detail below with reference to the accompanying drawings.

[0062] With reference to Figure 1 , Figure 2 and Figure 3 , the embodiment of the application provides a soil sampling device for soil organic carbon detection, which comprises an unmanned aerial vehicle body 1, a trundle assembly 2, a storage seat 3, a sampling cylinder 4, a driving mechanism 5, a transmission mechanism 6, a connecting rod 7, a transposition mechanism 8, a pushing mechanism 9, a clamping seat 10 and a monitoring mechanism.

[0063] The unmanned aerial vehicle body 1 is provided with the trundle assembly 2; the trundle assembly 2 can make the unmanned aerial vehicle body 1 walk on the ground; the unmanned aerial vehicle body 1 adopts a four-rotor or six-rotor multi-axle unmanned aerial vehicle, and the body material is light carbon fiber. A high-precision GPS positioning module (positioning error ≤0.5 m) and a millimeter wave radar are equipped, so that the unmanned aerial vehicle body 1 can be accurately hovered above a sampling point and avoid ground obstacles.

[0064] The unmanned aerial vehicle body 1 is fixedly provided below with the storage seat 3; a plurality of sampling cylinders 4 are slidably arranged in the storage seat 3; the sampling cylinder 4 is made of stainless steel 316L (corrosion resistant, avoiding metal ion pollution of soil organic carbon), and the inner wall is polished to reduce the sliding resistance of the sampling cylinder.

[0065] The unmanned aerial vehicle body 1 is fixedly provided with the driving mechanism 5; the connecting rod 7 is rotatably arranged on the driving mechanism 5.

[0066] The unmanned aerial vehicle body 1 is rotatably provided with the transmission mechanism 6; the connecting rod 7 is in transmission connection with the transmission mechanism 6; the connecting rod 7 is detachably fixedly connected with the sampling cylinder 4.

[0067] The storage seat 3 is fixedly provided with a transposition mechanism 8; the storage seat 3 is rotatably provided with a pushing mechanism 9; the pushing mechanism 9 and the transposition mechanism 8 are in transmission connection; and the pushing mechanism 9 is provided with a plurality of rubber plugs 95.

[0068] The storage seat 3 is fixedly provided with a clamping seat 10; and the clamping seat 10 is in transmission connection with the transposition mechanism 8.

[0069] The unmanned aerial vehicle body 1 is fixedly provided with a monitoring mechanism; the monitoring mechanism monitors the soil sampling process and timely discovers abnormal conditions.

[0070] In this technical scheme, in the initial state, the connecting rod 7 is fixedly connected with one of the sampling cylinders 4. The unmanned aerial vehicle body 1 flies to a position requiring sampling, starts the driving mechanism 5 to drive the sampling cylinder 4 to move downward, and drives the sampling cylinder 4 to rotate through the transmission mechanism 6, so that the sampling cylinder 4 extends into the ground, and the soil sampling operation is completed. Then, the driving mechanism 5 drives the sampling cylinder 4 to move upward to a predetermined position; the transposition mechanism 8 drives the pushing mechanism 9 to rotate, so that one of the rubber plugs 95 is below the sampling cylinder 4; the driving mechanism 5 drives the sampling cylinder 4 to move downward, so that the rubber plug 95 is clamped into the bottom of the sampling cylinder 4 to seal; then the driving mechanism 5 drives the sampling cylinder 4 to move upward by a distance, the connecting rod 7 is disconnected from the sampling cylinder 4. The transposition mechanism 8 drives the pushing mechanism 9 to rotate by 180 degrees, and drives the clamping seat 10 to rotate by a certain angle, so that the sampling cylinder 4 containing the soil sample is away from the position below the connecting rod 7, and the next sampling cylinder 4 is rotated to the position below the connecting rod 7 for standby. The monitoring mechanism monitors the soil sampling process and timely discovers abnormal conditions.

[0071] Further, the caster assembly 2 comprises an electric telescopic rod 21, a connecting seat 22, a sleeve 23 and a caster 24.

[0072] The unmanned aerial vehicle body 1 is fixedly provided with four machine legs 12; an electric telescopic rod 21 is fixedly arranged between the front two machine legs 12 and the rear two machine legs 12; the movable rod of the electric telescopic rod 21 is fixedly provided with the connecting seat 22; the sleeve 23 is fixedly arranged on the two sides of the connecting seat 22; the sleeve 23 is slidably arranged on the machine leg 12; and the caster 24 is rotatably arranged on the two sides of the connecting seat 22. The caster 24 is driven by electricity. A DC motor for driving the caster 24 to rotate is arranged on the connecting seat 22. A laser range finder is fixedly arranged below the connecting seat 22.

[0073] In this technical scheme, the electric telescopic rod 21 is started to drive the connecting seat 22 to move downward, so that the two groups of casters 24 contact the ground, and the machine legs 12 are lifted, and then the unmanned aerial vehicle body 1 can be moved on the ground for a short distance through the casters 24.

[0074] Reference Figure 2, the driving mechanism 5 includes a motor A51, a bevel gear A52, a spur gear A53 and a gear plate 54;

[0075] A motor A51 is fixedly provided on the drone body 1; a bevel gear A52 and a spur gear A53 are coaxially fixedly provided on the output end of the motor A51;

[0076] A guide tube 11 is fixedly provided on the drone body 1 , and a tooth plate 54 is slidably provided in the guide tube 11 . The tooth plate 54 is meshed and transmission-connected with the spur gear A53 .

[0077] The connecting rod 7 is rotatably arranged on the gear plate 54 through a bearing; the bevel gear A52 is transmission-connected to the transmission mechanism 6.

[0078] In this solution, starting motor A51 rotates bevel gear A52 and spur gear A53, which in turn drives toothed plate 54 downward, which in turn drives connecting rod 7 and sampling barrel 4 downward. Simultaneously, bevel gear A52 activates transmission mechanism 6, which rotates connecting rod 7 and sampling barrel 4. When the motor reverses, the toothed plate slides upward, releasing the sampling barrel 4 from the soil and lifting it back into storage receptacle 3.

[0079] Furthermore, the transmission mechanism 6 includes a gear column 61 and a bevel gear B62;

[0080] A gear column 61 is rotatably provided on the drone body 1; a bevel gear B62 is coaxially provided on the upper end of the gear column 61; and the bevel gear B62 is meshed and transmission-connected with the bevel gear A52.

[0081] A gear ring A71 is fixedly provided on the upper end of the connecting rod 7, and the gear ring A71 is meshed with the gear column 61 for transmission connection.

[0082] In this technical solution, the starting motor A51 drives the bevel gear A52 and the spur gear A53 to rotate, the bevel gear A52 drives the bevel gear B62 to rotate, the bevel gear B62 drives the gear column 61 to rotate, the gear column 61 drives the gear ring A71 and the connecting rod 7 to rotate, so that the connecting rod 7 rotates while moving downward, which is convenient for easy insertion into the soil and quick completion of the sampling operation.

[0083] Reference Figure 2 and Figure 5 , the shifting mechanism 8 includes a rotating seat 81, a spur gear B82, a spur gear C83 and a motor B84;

[0084] A rotating seat 81 is rotatably provided on the storage seat 3. An inner tooth groove 811 is provided on the inner side of the rotating seat 81. Two sets of driving teeth A812 are symmetrically provided on the outer lower part of the rotating seat 81. A set of driving teeth B813 is provided above one set of driving teeth A812.

[0085] The motor B84 is fixedly arranged on the storage seat 3, and the output end of the motor B84 is coaxially fixedly arranged with the spur gear B82; the spur gear C83 is rotatably arranged on the storage seat 3; and the spur gear C83 is in meshing transmission connection with the clamping seat 10. A plurality of tooth grooves are arranged on the inner side of the clamping seat 10;

[0086] The spur gear B82 is in meshing transmission connection with the inner tooth groove 811; and the spur gear C83 is in meshing transmission connection with the driving tooth B813.

[0087] The driving tooth A812 is in transmission connection with the pushing mechanism 9.

[0088] A plurality of arc-shaped clamping grooves 101 are uniformly arranged on the outer side of the clamping seat 10; and the arc-shaped clamping grooves 101 are matched with the sampling cylinder 4.

[0089] In this technical scheme, the motor B84 drives the spur gear B82 to rotate, the spur gear B82 drives the rotating seat 81 to rotate, the driving tooth A812 on the rotating seat 81 first drives the pushing mechanism 9 to rotate, so that the rubber plug 95 thereon is rotated to below the sampling cylinder 4 and the connecting rod 7, and the lower end of the sampling cylinder 4 is conveniently sealed; then the driving tooth A812 on the other side drives the pushing mechanism 9 to continue to rotate, so that the pushing mechanism 9 is separated from the position below the connecting rod 7, and the driving tooth B813 drives the clamping seat 10 to rotate by the spur gear C83, so that the next sampling cylinder 4 is below the connecting rod 7.

[0090] Referring to Figure 3 and Figure 4 , the pushing mechanism 9 comprises a cylinder 91, a tooth ring B92, a positioning pipe 93, a supporting seat 94, a rubber plug 95 and an electric push rod 97.

[0091] The positioning pipe 93 is fixedly arranged on the storage seat 3, and the cylinder 91 is rotatably arranged on the positioning pipe 93; a plurality of rubber plugs 95 are arranged in the cylinder 91, a top cover is detachably arranged on the top of the cylinder 91, and a spring is fixedly arranged below the top cover. The spring is in abutment with the uppermost rubber plug 95.

[0092] The tooth ring B92 is detachably fixedly arranged on the cylinder 91; and the tooth ring B92 is in meshing transmission connection with the driving tooth A812.

[0093] The supporting seat 94 is fixedly arranged at the lower end of the cylinder 91; the electric push rod 97 is fixedly arranged on the supporting seat 94, an active rod of the electric push rod 97 is fixedly arranged with a push plate, and the push plate is in abutment with the rubber plug 95.

[0094] The square hole 96 is arranged on the upper layer of one end of the supporting seat 94.

[0095] In this technical solution, the electric push rod 97 pushes the lowermost rubber plug 95 to the square hole 96; the driving tooth A812 drives the cylinder 91 to rotate, so that the square hole 96 and the rubber plug 95 are below the connecting rod 7, facilitating the installation of the rubber plug 95 to the lower end of the sampling cylinder 4. Then the rotating seat 81 continues to rotate, so that the supporting seat 94 and the square hole 96 are away from the position below the connecting rod 7. The movable rod of the electric push rod 97 retreats to the initial position, so that the lowermost rubber plug 95 falls in front of the push plate for standby.

[0096] Referring to Figure 1 and Figure 6 , the top of the sampling cylinder 4 is provided with a square groove 41; the lower end of the connecting rod 7 is fixedly provided with a square column, and an electromagnet is fixedly arranged in the square column. The size and shape of the square column match the square groove 41, the square column can be inserted into the square groove 41, and the sampling cylinder 4 and the connecting rod 7 are fixedly connected through the electromagnet. The square column is integrally formed with the connecting rod 7, and the square column and the square groove 41 are clearance fit, which ensures smooth insertion and no radial shaking. The annular electromagnet is embedded in the center of the square column, and a strong magnetic field is generated after electrification, which is tightly attracted to the ferromagnetic gasket of the square groove 41, preventing the sampling cylinder from falling off during sampling.

[0097] The bottom of the square groove 41 is provided with a ferromagnetic gasket (made of pure iron and plated with chromium for corrosion protection), which is magnetically attracted to the electromagnet of the connecting rod 7.

[0098] The top of the sampling cylinder 4 is provided with a ring groove 42, and a plurality of air guide holes 43 are arranged in the ring groove 42; a butyronitrile rubber sealing ring is arranged in the ring groove 42. When the square column is inserted into the square groove 41, the sealing ring is deformed by being pressed by the annular groove at the bottom of the connecting rod 7, realizing air path sealing.

[0099] The other end of the sampling cylinder 4 is provided with a plurality of air bag storage grooves 44, and an air bag is fixedly arranged in each air bag storage groove 44; the air guide holes 43 are in communication with the air bags. In the initial state, the air bag is attached to the air bag storage groove 44, and the outer side is aligned with the inner wall of the sampling cylinder 4. The air bag is made of food-grade silicone material and is in the shape of a long strip, one end of which is fixed to the groove bottom by high-strength glue, and the other end is a free end; in the initial state, it is attached to the groove, and after inflation, the air bag gathers towards the center to form an annular sealing surface, completely sealing the soil in the cylinder.

[0100] The bottom of the connecting rod 7 is provided with an annular groove;

[0101] A gas pump is fixedly arranged on the unmanned aerial vehicle body 1, and the output end of the gas pump is in communication with the annular groove at the bottom of the connecting rod 7. The output end of the gas pump is fixedly provided with a gas pressure gauge. A micro diaphragm air pump (model FML200) is selected and fixed in the equipment cabin of the unmanned aerial vehicle body 1, and the output end is connected with the rotary joint of the connecting rod 7 through a pressure-resistant silicone tube. The output end of the air pump is connected in series with a digital air pressure gauge (model BMP280) to monitor the air pressure in the air bag in real time; the main controller sets the inflation threshold, and the air pump stops when the air pressure reaches the threshold, and maintains the pressure after sealing.

[0102] Preferably, a plurality of sawteeth are arranged below the connecting rod 7.

[0103] In this technical solution, after the square column on the connecting rod 7 is inserted into the square groove 41, the electromagnet works, and the electromagnet fixes the sampling cylinder 4 and the connecting rod 7. After the soil sampling is completed, the air bag is inflated by the air pump, so that the air bag is unfolded, and the plurality of air bags seal the lower end of the sampling cylinder 4 to prevent the sample from falling off.

[0104] Further, the monitoring mechanism comprises:

[0105] The data collection module collects geological data and vegetation data of the soil sampling area; acquires regional soil types (such as black soil, red soil), parent material types, slope, and soil thickness through satellite remote sensing; identifies vegetation types (forest / farmland / grassland), coverage, and phenology (such as crop growth stage), and correlates to the soil organic carbon input model. Historical data is collected, and the data is labeled.

[0106] The route planning module reasonably plans the data and position of the sampling points according to the geological data and sampling requirements of the soil sampling area, and plans the optimal flight and walking route of the unmanned aerial vehicle body 1. Flight is adopted for long distances, and the foot wheel assembly 2 is adopted for short distances and under the condition of adjustment permission. The soil humidity sensor is fixedly arranged on the unmanned aerial vehicle body 1;

[0107] The image acquisition module includes a plurality of high-definition cameras for collecting high-definition images of the soil sampling process; two macro cameras are installed on both sides of the unmanned aerial vehicle body, and a 120° wide-angle lens is used to take pictures of the environment within a five-meter range around the sampling point (such as vegetation damage and human interference marks) at 30 frames per second; two device cameras are installed on the outer side of the storage seat 3 at an angle of 45°, focusing on the whole process of the sampling cylinder 4 entering the soil and lifting, and capturing the wear and soil adhesion of the sampling cylinder;

[0108] The image analysis module analyzes and identifies the collected images in a timely manner to identify abnormal conditions in the soil sampling process;

[0109] Ground penetrating radar (GPR): The ground penetrating radar is fixedly arranged on the unmanned aerial vehicle body 1, emits high-frequency electromagnetic waves, identifies foreign objects through the difference in electromagnetic characteristics of different media, and can generate underground profile images. It has good recognition effect on foreign objects such as stones, roots, and pipelines, and is suitable for scanning the region before sampling to plan to avoid foreign object dense areas.

[0110] The weather data collection module collects real-time weather data of the soil sampling area;

[0111] The radar data analysis module analyzes the radar data to identify underground foreign objects such as stones, roots, and pipelines in the soil sampling area in a timely manner.

[0112] Alarm module: including alarm, when monitoring abnormal situation, timely alarm.

[0113] Control center: with data collection module, route planning module, image acquisition module, image analysis module, ground penetrating radar, radar data analysis module and alarm module.

[0114] Further, the image analysis module analyzes and identifies the collected images, and identifies abnormal conditions in the soil sampling process in time; including the following steps:

[0115] 1. Image preprocessing: the original images collected by different cameras are uniformly preprocessed to eliminate interference factors: including format standardization, noise filtering and region of interest (ROI) cropping.

[0116] Format standardization: convert all images to RGB format, unify frame rate, keep key frames for static images, and take five key frames per second for dynamic video.

[0117] Noise filtering: for uneven brightness caused by outdoor strong light / shadow, use adaptive histogram equalization algorithm to enhance contrast; for blurring caused by camera shaking, use non-local mean denoising algorithm to repair; for macro images of sample camera, use Gaussian filter to remove sensor noise and highlight soil texture details.

[0118] Region of interest (ROI) cropping: automatically crop the sampling cylinder blade (ROI area is the lower 1 / 3 part of the sampling cylinder), rubber plug and other parts;

[0119] 2. Multi-dimensional feature extraction: for different monitoring targets, extract exclusive feature parameters to provide data support for subsequent identification:

[0120] 2.1. Device state feature extraction:

[0121] 2.1.1. Sampling cylinder integrity: use Canny operator to extract the sampling cylinder blade contour, calculate the contour continuity, and if there is a break, mark it as "blade damage feature";

[0122] Deformation analysis, through Hough transform to detect the central axis of the sampling cylinder, and compare with the standard template, if the bending degree is > 3°, mark it as "cylinder deformation feature".

[0123] 2.1.2. Connection and sealing components:

[0124] Square slot-square column fit: extract the image of square column inserted into square slot, calculate the gap uniformity, normal gap difference ≤0.1mm, if local gap >0.3mm, mark it as connection loose feature;

[0125] Rubber plug position: Threshold segmentation of the plug area by color, calculate the deviation of its center from the sampling cylinder axis, normal ≤1mm, deviation >2mm is marked as "plug misalignment feature".

[0126] 2.1.3, caster and walking parts: Optical flow analysis is performed on consecutive frame images to calculate the caster rotation speed. If the speed is 0 and the machine body has displacement, it is marked as a caster jam feature.

[0127] 2.2, sample state feature extraction:

[0128] 2.2.1, soil filling integrity:

[0129] Volume calculation: Through the sample camera image, use binocular vision algorithm to reconstruct the three-dimensional model of the soil in the sampling cylinder, calculate the filling rate, normal should be ≥85%, <60% is marked as insufficient sample amount feature;

[0130] Impurity identification: Use color clustering and shape analysis, impurities are mostly irregular polygons, count the impurity ratio, normal ≤5%, >10% is marked as impurity exceeding standard feature.

[0131] 2.2.2, soil layering integrity: Extract the internal layering lines of the soil, calculate the number of line breaks, too many are marked as layering damage features caused by sampling cylinder jam or foreign matter interference.

[0132] 2.3, environmental interference feature extraction:

[0133] External obstacles: Use YOLOv8 model to identify pedestrians, animals, large stones in macro images, marked as "external interference feature";

[0134] Vegetation interference: Use U-Net model to segment the vegetation area in the image, if the vegetation covers the sampling point center 30cm range and the height >10cm, mark it as a vegetation obstruction feature.

[0135] 3, accurate identification of abnormal types: Based on the trained deep learning model, classify and identify the extracted features, output the abnormal type and confidence:

[0136] The model architecture uses a dual-branch fusion model; one branch processes static features (such as shape, color) using a CNN network (ResNet50); the other branch processes dynamic features using an LSTM network; finally, the results are fused through an attention mechanism to improve recognition accuracy.

[0137] Abnormal type identification:

[0138] Device abnormalities (such as blade cracking, plug misalignment, caster jam): The model outputs the specific components and the degree of abnormality.

[0139] Sample abnormalities (e.g. excessive impurities, layer damage): output abnormal indicators;

[0140] Environmental abnormalities (e.g. pedestrians approaching, vegetation blocking): output interference sources and distances.

[0141] 4. Time series dynamic comparison: predict potential abnormal trends; through the analysis of feature changes in consecutive multiple images, predict potential abnormalities that have not yet occurred but may occur:

[0142] Trend calculation:

[0143] Blade wear: continuously calculate the missing area of the blade profile, if it increases from 5% to 20% (speed > 5% / 10 seconds), predict "possible serious collapse within one minute";

[0144] Soil adhesion: count the pixel ratio of soil adhesion on the outer wall of the sampling cylinder, if it increases from 10% to 30% (speed > 10% / 10 seconds), predict "risk of sampling cylinder blockage increases";

[0145] Wheel slip: calculate the ratio of wheel rotation angle to body displacement, if the ratio of three consecutive frames is <0.5, predict "wheel is about to sink into soft soil".

[0146] Threshold triggering: when the trend value exceeds the preset threshold, generate a potential abnormality warning.

[0147] 5. Abnormal classification and decision output: according to the severity of the abnormality, the results are divided into three levels, and the corresponding processing instructions are output:

[0148] First-level abnormality: does not affect core sampling; sends "prompt information" to the control center, does not interrupt the sampling process.

[0149] Second-level abnormality: affects sample quality or light damage to equipment; send warning information, control center suspends current sampling, starts correction program.

[0150] Third-level abnormality: may cause equipment damage or sample completely invalid; send "emergency alert", control center triggers alarm module, and starts emergency program.

[0151] 6: Data feedback: bind the abnormality recognition result with the corresponding sampling point ID and store it in the database for subsequent traceability analysis.

[0152] In this technical solution, through the dual mechanism of static feature recognition and dynamic trend prediction, this process not only can discover the abnormality that has occurred in real time, but also can early warning potential risks, shorten the response time of abnormality handling, effectively reduce the equipment failure rate and invalid sample rate, provide reliable guarantee for the continuity of soil sampling and sample quality.

[0153] Further, the radar data analysis module analyzes the radar data and timely identifies the underground foreign matter in the soil sampling area. The steps include:

[0154] 1. Raw radar data preprocessing: including data format conversion and noise filtering;

[0155] Data format conversion: convert the original electromagnetic signal collected by the ground penetrating radar (GPR) into a two-dimensional matrix format for analysis, retaining key parameters such as signal intensity and propagation time.

[0156] Noise filtering: including removing system noise, suppressing environmental interference and correcting signal attenuation;

[0157] Remove system noise: eliminate the electronic noise of the radar device itself through smoothing filtering (such as sliding window average), and highlight the effective signal;

[0158] Suppress environmental interference: use background subtraction method to subtract the average signal of the area without foreign matter, and weaken the surface clutter, electromagnetic interference, etc.

[0159] Correct signal attenuation: due to the absorption of electromagnetic waves by soil medium, deep layer signal will be attenuated, through gain compensation, to ensure that the signal intensity at different depths can be compared. The signal gain compensation model is:

[0160] G(d, w) = G0·exp[k0(1+0.02w)d]; In the formula, G(d, w) represents the signal gain coefficient at the detection depth d and the soil volume moisture content w, which is used to compensate for the signal attenuation caused by depth and humidity during the propagation of electromagnetic waves in soil, ensuring that the signal intensity under different conditions can be compared. G0 is the initial gain at the surface, the default value is 1.0, which is the initial amplification coefficient of electromagnetic waves at the surface, serving as the basis for gain calculation in the entire depth range. k0 is the basic attenuation coefficient, reflecting the basic rate of electromagnetic wave attenuation with depth in pure dry soil: k0 = 0.015 in clay soil (clay has strong absorption of electromagnetic waves, and the basic attenuation rate is higher); k0 = 0.008 in sandy soil (sandy soil has good permeability and weak absorption of electromagnetic waves, and the basic attenuation rate is lower). w is the soil volume moisture content, i.e. the percentage of water volume in soil total volume, which is collected by the soil moisture sensor (such as TDR sensor) carried by the unmanned aerial vehicle for real-time dynamic correction of attenuation coefficient; the higher the humidity, the stronger the absorption of electromagnetic waves by soil, and the attenuation coefficient needs to be increased accordingly. d is the detection depth, indicating the vertical depth position of the underground to be detected, the greater the depth, the more significant the attenuation of electromagnetic waves, and the signal needs to be amplified by gain coefficient.

[0161] 2. Radar profile image generation: including time-depth conversion, gray-scale color imaging and image enhancement;

[0162] Time-depth conversion: convert the travel time of electromagnetic signals (unit: nanoseconds) into the actual probing depth (unit: centimeters), calculated based on the electromagnetic wave propagation velocity in soil. The time-depth conversion model is:

[0163] D = [V0(1-0.01p)t] / [2(1+0.03w)]; where D is the underground depth (unit: cm), i.e. the actual vertical burial depth of the underground target detected by the electromagnetic wave. V0 is the propagation velocity of electromagnetic waves in pure dry soil (unit: cm / ns), which is a basic speed parameter and is set differently according to soil types: V0 = 12 in clay soil; V0 = 18 in sandy soil. p is the soil bulk density, i.e. the dry weight per unit volume of soil, reflecting the degree of soil compaction (calculated by inverse calculation of the resistance to entering the soil collected by the sampling cylinder pressure sensor). t is the round-trip travel time of electromagnetic wave signals, i.e. the total time from radar signal transmission to receiving reflected signals (recorded directly by the radar equipment). w is the volumetric water content of the soil, i.e. the proportion of water volume in the soil to the total volume of the soil (real-time collected by the soil moisture sensor carried by the unmanned aerial vehicle).

[0164] Gray color imaging: map the signal intensity to image pixel values to form the underground profile image; areas with strong signals (such as foreign object reflections) are displayed as bright spots / warm colors, and areas with weak signals (such as uniform soil) are displayed as dark spots / cold colors.

[0165] Image enhancement: strengthen the boundary features of foreign objects and surrounding soil through contrast stretching, edge sharpening, etc., such as the circular edges of stones and the linear profiles of pipes.

[0166] 3. Abnormal area identification: including threshold segmentation, region growing and artifact removal;

[0167] Threshold segmentation: set a signal intensity threshold, mark areas in the image with signal intensity exceeding the threshold as "abnormal areas", which may contain foreign objects.

[0168] Region growing: morphologically process the initially marked abnormal areas, merge adjacent discrete abnormal points to form complete foreign object candidate areas.

[0169] Artifact removal: including exclusion of surface reflection interference and differentiation of soil layering reflections;

[0170] Exclusion of surface reflection interference: strong reflection will occur at the interface between the surface and the air, which is removed by depth range limitation;

[0171] Differentiation of soil layering reflections: normal soil layering reflections appear as continuous parallel stripes, if the abnormal areas appear as irregular blocks or lines, they are determined as potential foreign objects.

[0172] 4. Foreign object feature extraction: for each candidate abnormal area, extract morphological and signal features to establish classification criteria:

[0173] 4.1, Morphological characteristics:

[0174] Stone: Mostly irregular blocks, with round, square or polygonal outlines, horizontal dimensions usually five to thirty centimeters, and vertical depth ranges are relatively concentrated;

[0175] Root: Linear or branched, irregular outline, long horizontal extension, moderate signal strength;

[0176] Metal pipe: Continuous linear feature, extremely strong signal strength, long and straight horizontal extension, and uniform diameter.

[0177] 4.2, Signal characteristics:

[0178] Reflection intensity: Metal pipe > stone > root > soil;

[0179] Signal attenuation: Roots absorb electromagnetic waves strongly, and signal attenuation is fast; stone attenuation is slow, and deep layers can still be detected;

[0180] Waveform characteristics: Pipe reflection signals are periodic, stone reflections are single-peak strong signals, and root reflections are multi-peak weak signals.

[0181] 5, Foreign object type determination:

[0182] 5.1, Template matching: Establish a feature template library of known foreign objects (including typical morphologies and signal characteristics of stone, root, and pipe), compare the features of the candidate area with the template, and calculate the similarity.

[0183] 5.2, Application of classification rules:

[0184] If the candidate area is "strong reflection + block + no branch", it is determined to be "stone";

[0185] If it is "medium reflection + linear / branch + signal attenuation fast", it is determined to be "root";

[0186] If it is "extremely strong reflection + continuous linearity + uniform diameter", it is determined to be "metal pipe";

[0187] For areas that cannot be matched, mark them as "unknown foreign objects" and need to be manually reviewed.

[0188] 6, Three-dimensional positioning coordinate output:

[0189] 6.1, Spatial coordinate conversion: Combine the GPS positioning data of the unmanned aerial vehicle and the horizontal distance and depth detected by the radar to calculate the three-dimensional coordinates of the foreign object.

[0190] 6.2, Size estimation: Estimate the actual size of the foreign object based on the pixel size of the abnormal area in the image and the scale.

[0191] 6.3, Result arrangement: generate a foreign matter list containing the type, coordinates, size, and confidence of each foreign matter.

[0192] 7: Support sampling decision: send the foreign matter list to the route planning module to automatically avoid foreign matter dense areas and adjust the sampling point position; push high-risk foreign matter information to the control center to prompt enhanced device protection during sampling; store foreign matter data in the database for subsequent analysis.

[0193] In this technical solution, accurate identification of underground foreign matters is achieved, especially distinguishing between different types such as stones, tree roots, and pipelines, providing an obstacle avoidance guide for soil sampling, reducing equipment wear and tear and invalid sampling, and improving the safety and efficiency of automated sampling.

[0194] Further, the route planning module reasonably plans the data and position of the sampling points based on the geological data and sampling requirements of the soil sampling area, and plans the optimal flight and walking route of the unmanned aerial vehicle main body 1, including the following steps:

[0195] 1, Multi-source basic data integration:

[0196] 1.1, Core data collection: call the geological data (soil type, slope, soil thickness) provided by the data collection module, vegetation data (coverage, type), and historical sampling records; obtain the preset sampling requirement parameters (such as the number of sampling points, target depth, and regional range); access the real-time data of the soil moisture sensor carried by the unmanned aerial vehicle (record the soil volume moisture content w of the 0-5cm surface layer in the sampling area).

[0197] 1.2, Data preprocessing: superimpose the geological data and humidity data on the electronic map to generate a soil humidity zoning map (divided into dry area w≤15%, moderate humidity area 15%<w≤30%, and high humidity area w>30% according to w value);

[0198] Mark the known obstacle areas (such as foreign matter dense areas detected by radar and slopes >15) as "no sampling / no walking areas".

[0199] 2, Preliminary layout of sampling points:

[0200] 2.1, Selection of point layout method: for homogeneous areas (such as plain farmland), use grid point layout method: generate initial sampling points on the map at a preset interval;

[0201] For heterogeneous areas (such as mountainous areas and forest land), use hierarchical random point layout method: generate sampling points randomly in each sub-area according to soil type and vegetation zoning.

[0202] 2.2, Sampling point screening: remove the initial points located in the no-sampling area and replace them with the nearest feasible area; ensure that the distance between adjacent sampling points is not less than the preset minimum value;

[0203] Priority should be given to retaining sampling points within the suitable wet zone (15%<w≤30%), because the soil structure in this area is stable, which is convenient for walking with casters and can reduce soil adhesion or collapse during sampling.

[0204] 3. Feasibility verification of sampling points:

[0205] 3.1. Single-point suitability assessment: For each preliminary sampling point, evaluate the soil moisture and topography:

[0206] Dry area: If the slope is ≤10°, it is determined to be a feasible area for casters;

[0207] Suitable wet area: If the slope is ≤15°, it is determined to be a caster priority area;

[0208] High humidity area: It is determined to be a flight area. Casters are prone to slipping or getting stuck in mud, and drones must be flown to reach it.

[0209] 3.2. Abnormal point correction: If a sampling point is located in a high humidity area but sampling is required, it will be marked as a mandatory sampling point and the subsequent route planning will be based on flight arrival.

[0210] If three adjacent sampling points are all high-humidity areas, appropriately reduce the number of sampling points in this area to avoid ineffective operations, and record the reasons.

[0211] 4. Segment planning of the route:

[0212] 4.1. Grouping and sorting of sampling points: Adopt the principle of clustering close to each other: group sampling points with close distance (≤100m) into the same group to reduce long-distance round trips;

[0213] The order of visiting sampling points in each group is sorted according to the principle of shortest straight-line distance.

[0214] 4.2、Flying and walking mode division:

[0215] Flight mode triggering: When the distance between two groups of sampling points is greater than 50m, plan the drone's flight route (fly in a straight line and avoid obstacles); if the sampling points within a single group are located in a high-humidity area or a prohibited area (such as a steep slope), use the flight mode to fly back and forth; the flight route must stay away from areas with vegetation height greater than 0.5m (to avoid propeller entanglement).

[0216] Walking mode triggering: The distance between sampling points in a single group is ≤50m and all are located in the caster feasible area / priority area (ω≤30%, slope ≤15°). The caster walking route is planned; the walking route gives priority to paths with lower humidity in dry areas or suitable humidity areas, avoiding the boundaries of high humidity areas to prevent casters from slipping suddenly when entering high humidity areas from suitable humidity areas.

[0217] 5. Dynamic route optimization:

[0218] 5.1, Real-time humidity feedback adjustment: During the flight of the unmanned aerial vehicle, the soil humidity sensor updates the humidity data in the range of 5m in front of the vehicle in real time; if a local high-humidity area appears in the original planned route, the vehicle automatically switches to short-distance flight or detours to the adjacent dry area.

[0219] 5.2, Energy consumption balance optimization: calculate the energy consumption of different modes, the flight mode is estimated by distance x 0.1Wh / m, and the walking mode is estimated by distance x 0.03Wh / m+humidity coefficient;

[0220] For long-distance routes, if the flight energy consumption is less than the walking energy consumption, the flight mode is automatically selected; for short-distance routes, the walking mode is automatically selected.

[0221] 5.3, Emergency backup route generation: one or two backup routes are planned for each group of sampling points, and when the main route is not available due to sudden humidity changes or temporary obstacles, the backup route is immediately switched to.

[0222] 6, Route output: generate the final route scheme, output the detailed route map containing the sampling point coordinates, access order, and mode switching points; mark the key parameters such as the humidity range of each walking segment, the altitude of the flight segment, and the estimated total time. The route scheme is sent to the unmanned aerial vehicle control system for real-time synchronization of position information; when walking, the speed instruction is sent to the wheel assembly;

[0223] The application provides a soil sampling method for soil organic carbon detection, comprising the following steps:

[0224] S1, the data collection module of the monitoring mechanism collects geological data and vegetation data of the area where soil sampling is needed; historical data is collected and labeled.

[0225] S2, the route planning module reasonably plans the data and position of the sampling points according to the geological data of the soil sampling area and the sampling requirements, and plans the optimal flight and walking route of the unmanned aerial vehicle body 1;

[0226] S3, the unmanned aerial vehicle body 1 carries the ground penetrating radar GPR to scan the sampling area and plans to avoid the area with dense foreign matters; the radar data analysis module analyzes the radar data and timely identifies the foreign matters under the soil sampling area; the meteorological data collection module collects the meteorological data of the soil sampling area in real time;

[0227] S4, the route planning module adjusts the flight and walking route according to the analysis result of the radar data analysis module;

[0228] S5, the unmanned aerial vehicle body 1 drives the sampling cylinder 4 to perform soil sampling operation through the driving mechanism 5, the transmission mechanism 6, the connecting rod 7, the transposition mechanism 8, the pushing mechanism 9 and the clamping seat 10.

[0229] S6, the image acquisition module collects high-definition images of the soil sampling process;

[0230] S61, initial state, the connecting rod 7 is fixedly connected to one of the sampling tubes 4. The drone body 1 flies to the location where sampling is required;

[0231] S62, start the driving mechanism 5 to drive the sampling tube 4 to move downward, and at the same time drive the sampling tube 4 to rotate through the transmission mechanism 6, so that the sampling tube 4 extends into the ground to perform soil sampling operations.

[0232] S62, the driving mechanism 5 drives the sampling tube 4 to move upward to a predetermined position;

[0233] S63, the shifting mechanism 8 drives the pushing mechanism 9 to rotate, so that one of the rubber plugs 95 is located below the sampling tube 4;

[0234] S64, the driving mechanism 5 drives the sampling tube 4 to move downward, so that the rubber plug 95 is stuck into the bottom of the sampling tube 4 to perform a sealing operation;

[0235] S65 , the driving mechanism 5 then drives the sampling cylinder 4 to move upward for a certain distance, and the connecting rod 7 cancels the connection with the sampling cylinder 4 .

[0236] S66, the shifting mechanism 8 drives the pushing mechanism 9 to rotate 180 degrees, and at the same time drives the holder 10 to rotate a certain angle, so that the sampling tube 4 containing the soil sample leaves the position under the connecting rod 7, and the next sampling tube 4 rotates to the bottom of the connecting rod 7 for standby.

[0237] S7, the image analysis module analyzes and identifies the collected images and promptly identifies abnormal conditions during the soil sampling process;

[0238] S8. When an abnormal situation is detected, the alarm module will issue an alarm in time.

[0239] In view of the fact that soil organic carbon is easily disturbed, the present invention adopts a design such as sealing the inert material sampling tube with a rubber plug during the sampling process to reduce the contact between the sample and the outside world, avoid volatilization or contamination of organic carbon, and ensure the original state of the sample before detection. The route planning module combines a hybrid mode of drone flight and caster walking, which can quickly reach long-distance flights and save energy for short-distance walking. Compared with traditional manual sampling or pure flight mode, the operation efficiency is improved and more sampling points can be completed in a single operation. The transposition mechanism realizes automatic replacement and sealing of the sampling tube without manual intervention, shortens the sampling cycle of a single group of samples, and is suitable for batch sampling needs in large areas. The ground penetrating radar and radar data analysis module scan underground foreign objects in advance, and the route planning module dynamically adjusts the path. The whole process is automated and intelligent, reducing human errors.

[0240] The above-described embodiments are only used to illustrate the technical solutions of the present application, and are not intended to limit the present application; although the present application has been described in detail with reference to the foregoing embodiments, it should be understood by those skilled in the art that the technical solutions recorded in the foregoing embodiments can still be modified, or some technical features can be replaced by equivalent replacements; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A soil sampling method for detecting soil organic carbon, characterized in that: The following steps are involved: S1. The data collection module of the monitoring agency collects geological data and vegetation data and labels the data; S2, the route planning module rationally plans the data and location of the sampling points based on the geological data and sampling requirements of the soil sampling area, and plans the optimal flight and walking route of the drone body; S3. The drone carries a ground-penetrating radar to scan the sampling area. The radar data analysis module analyzes the radar data to identify foreign objects underground in the soil sampling area. The meteorological data collection module collects meteorological data in real time. S4. The route planning module adjusts the flight and walking routes according to the analysis results of the radar data analysis module; S5. The drone body uses the driving mechanism, transmission mechanism, connecting rod, transposition mechanism, pushing mechanism and card seat to make the sampling tube perform soil sampling operation; S6, the image acquisition module collects high-definition images of the soil sampling process; S7, the image analysis module analyzes and identifies the collected images and identifies abnormal conditions in the soil sampling process; S8. When an abnormal situation is detected, the alarm module will issue an alarm in time.

2. The soil sampling method for detecting soil organic carbon according to claim 1, characterized in that: Step S6 includes the following steps: S61, in the initial state, the connecting rod is fixedly connected to one of the sampling tubes; the drone body flies to the location where sampling is required; S62, starting the driving mechanism to drive the sampling tube to move downward, and at the same time driving the sampling tube to rotate through the transmission mechanism, so that the sampling tube extends into the ground to perform soil sampling; S62, the driving mechanism drives the sampling tube to move upward to a predetermined position; S63, the shifting mechanism drives the pushing mechanism to rotate so that one of the rubber plugs is located below the sampling tube; S64, the driving mechanism drives the sampling tube to move downward, so that the rubber plug is stuck into the bottom of the sampling tube to perform a sealing operation; S65, the driving mechanism drives the sampling tube to move upward for a certain distance, and the connecting rod is disconnected from the sampling tube; S66. The position-changing mechanism drives the pushing mechanism to rotate 180 degrees, and at the same time drives the holder to rotate a certain angle, so that the sampling tube containing the soil sample leaves the position under the connecting rod, and the next sampling tube rotates to the bottom of the connecting rod for standby.

3. The soil sampling method for detecting soil organic carbon according to claim 1, characterized in that: Step S7 includes the following steps: S71, Image preprocessing: Preprocess the collected images, including format standardization, noise filtering and region of interest cropping; S72. Multi-dimensional feature extraction: Extract unique feature parameters for different monitoring targets, including equipment status feature extraction, sample status feature extraction, and environmental interference feature extraction; S73. Accurately identify anomaly types: Based on the trained deep learning model, the extracted features are classified and identified, and the anomaly type and confidence level are output: S74, Time Series Dynamic Comparison: Predict potential abnormal trends; by analyzing the characteristic changes of multiple consecutive frames of images, predict abnormalities that have not yet occurred but may occur: S75, abnormal classification and decision output; S76: Data feedback: Bind the abnormality identification result with the corresponding sampling point ID and store it in the database.

4. The soil sampling method for detecting soil organic carbon according to claim 1, wherein: Step S3 includes the following steps: S31. Raw radar data preprocessing: including data format conversion and noise filtering; S32, Radar profile image generation: including time-to-depth conversion, grayscale color imaging and image enhancement; S33, abnormal region identification: including threshold segmentation, region growing and artifact removal; S34, foreign body feature extraction: for each candidate abnormal area, extract the morphology and signal features and establish the classification basis; S35, foreign body type determination; S36, 3D positioning coordinate output: including spatial coordinate conversion, size estimation and result collation; S37: Support sampling decisions: Send the foreign object list to the route planning module, automatically avoid areas with dense foreign object concentrations, and adjust the sampling point location; push high-risk foreign object information to the control center.

5. The soil sampling method for detecting soil organic carbon according to claim 2, characterized in that: The caster assembly includes an electric telescopic rod, a connecting seat, a sleeve and a caster; Four legs are fixedly arranged under the main body of the drone; an electric telescopic rod is fixedly arranged between the two front legs and the two rear legs, a connecting seat is fixedly arranged on the movable rod of the electric telescopic rod, and sleeves are fixedly arranged on both sides of the connecting seat; the sleeves are slidably arranged on the legs; casters are rotatably arranged on both sides of the connecting seat; The driving mechanism includes a motor A, a bevel gear A, a spur gear A and a gear plate; A motor A is fixedly installed on the drone body; a bevel gear A and a spur gear A are coaxially fixedly installed on the output end of the motor A; a guide tube is fixedly installed on the drone body, and a toothed plate is slidably installed in the guide tube, and the toothed plate is meshed and connected to the spur gear A; the bevel gear A is connected to the transmission mechanism.

6. The soil sampling method for detecting soil organic carbon according to claim 5, characterized in that: The transmission mechanism includes a tooth column and a bevel gear B; A gear column is rotatably provided on the main body of the drone; a bevel gear B is coaxially provided on the upper end of the gear column; bevel gear B is meshed and connected with bevel gear A; a gear ring A is fixedly provided on the upper end of the connecting rod, and the gear ring A is meshed and connected with the gear column; The shifting mechanism includes a rotating seat, a spur gear B, a spur gear C, and a motor B. A rotating seat is rotatably provided on the storage seat, and an inner tooth groove is provided on the inner side of the rotating seat. Two sets of driving teeth A are symmetrically provided on the lower outer side of the rotating seat. A set of driving teeth B is provided above one set of driving teeth A. A motor B is fixedly installed on the storage base, and a spur gear B is coaxially fixedly installed on the output end of motor B; a spur gear C is rotatably installed on the storage base; the spur gear C is meshed and connected to the card base; a number of tooth grooves are provided on the inner side of the card base; the spur gear B is meshed and connected to the inner tooth grooves; the spur gear C can be meshed and connected to the drive gear B; the drive gear A can be connected to the pushing mechanism; a number of arc-shaped slots are evenly distributed on the outer side of the card base; the arc-shaped slots are adapted to the sampling tube.

7. The soil sampling method for detecting soil organic carbon according to claim 6, characterized in that: The pushing mechanism includes a cylinder, a gear ring B, a positioning tube, a supporting seat, a rubber plug and an electric push rod; A positioning tube is fixedly provided on the storage seat, and a cylinder is rotatably provided on the positioning tube; a number of rubber plugs are provided in the cylinder, and a gear ring B is detachably fixedly provided on the cylinder; the gear ring B can be engaged with the driving tooth A for transmission connection; a supporting seat is fixedly provided on the lower end of the cylinder; an electric push rod is provided on the supporting seat, and a push plate is fixedly provided on the movable rod of the electric push rod, and the push plate abuts against the rubber plug.

8. The soil sampling method for detecting soil organic carbon according to claim 7, characterized in that: A square groove is provided on the top of the sampling tube; a square column is fixedly provided on the lower end of the connecting rod, and an electromagnet is fixedly provided in the square column. The size and shape of the square column match the square groove, and a ferromagnetic gasket is provided at the bottom of the square groove to form a magnetic attraction with the electromagnet of the connecting rod; an annular groove is provided on the top of the sampling tube, and a number of air guide holes are provided in the annular groove; a nitrile rubber sealing ring is built into the annular groove; a number of air bag storage grooves are opened at the other end of the sampling tube, and air bags are fixedly provided in the air bag storage grooves; the air guide holes are connected with the air bags; an annular groove is provided at the bottom of the connecting rod; an air pump is fixedly provided on the main body of the drone, and the output end of the air pump is connected with the annular groove at the bottom of the connecting rod; a barometer is fixedly provided at the output end of the air pump, and a number of serrations are provided under the connecting rod.

9. The soil sampling method for detecting soil organic carbon according to claim 1, wherein: Monitoring agencies include: Data collection module: collect geological data and vegetation data of areas requiring soil sampling; collect historical data and annotate the data; Route planning module: Based on the geological data and sampling requirements of the soil sampling area, the data and location of the sampling points are rationally planned, and the optimal flight and movement routes of the drone are planned; Image acquisition module: includes multiple high-definition cameras to collect high-definition images of the soil sampling process; Image analysis module: Analyze and identify the collected images to identify abnormal conditions during soil sampling; Ground-penetrating radar: A ground-penetrating radar is fixed on the main body of the drone, emitting high-frequency electromagnetic waves to identify foreign objects based on the differences in electromagnetic properties of different media; Meteorological data collection module: collects meteorological data of the soil sampling area in real time; Radar data analysis module: Analyzes radar data and identifies foreign objects underground in the sampling area; Alarm module: including alarm, which will issue an alarm in time when any abnormal situation is detected; Control center: data collection module, route planning module, image acquisition module, image analysis module, ground penetrating radar, radar data analysis module and alarm module.

10. A soil sampling device for detecting soil organic carbon, comprising: UAV body, caster assembly, storage seat, sampling tube, driving mechanism, transmission mechanism, connecting rod, transposition mechanism, pushing mechanism, card seat and monitoring mechanism; characterized by: The drone body is provided with a caster assembly; a storage seat is fixedly provided below the drone body; a plurality of sampling cartridges are slidably provided in the storage seat; a driving mechanism is fixedly provided on the drone body; a connecting rod is rotatably provided on the driving mechanism; a transmission mechanism is rotatably provided on the drone body; the connecting rod is in transmission connection with the transmission mechanism; the connecting rod is detachably fixedly connected to the sampling cartridge; A transposition mechanism is fixedly provided on the storage seat; a pushing mechanism is rotatably provided on the storage seat; the pushing mechanism and the transposition mechanism are transmission connected; a plurality of rubber plugs are provided on the pushing mechanism; a card seat is fixedly provided on the storage seat; the card seat is transmission connected to the transposition mechanism; and a monitoring mechanism is provided on the drone body.

Images