Soil particle mesoscopic parameter calibration method and device and unmanned vehicle simulation method

Abstract

The invention discloses a soil particle mesoscopic parameter calibration method which is characterized by comprising the following steps: step 1, applying a predetermined load to a soil sample, and obtaining an actual soil settlement amplitude under the predetermined load; step 2, acquiring a soil sample image and extracting geometric parameters of soil particles; 3, soil particle mesoscopic parameters are input; 4, according to the actual soil settlement amplitude in the step 1, the soil particle geometric parameters in the step 2 and the soil particle mesoscopic parameters input in the step 3, calculating the soil particle mesoscopic parameters through an iterative optimization algorithm; and step 5, outputting the calculated soil particle mesoscopic parameters as calibrated soil mesoscopic parameters, and through the high-resolution image and the iterative optimization algorithm, the accuracy of soil particle mesoscopic parameter calibration is improved, and the calibration precision is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned vehicle mobility simulation evaluation, and in particular to a soil particle microscopic parameter calibration method and device, and an unmanned vehicle simulation method. Background Art

[0002] In the field of unmanned vehicles, especially when they are used on soft ground (such as deserts, swamps, snow, etc.), it is crucial to ensure that the vehicle has good passability and stability. Traditionally, the mobility assessment of vehicles in these complex terrains mainly relies on macroscopic mechanical models, which are usually based on the overall properties of the soil, such as hardness, compression modulus and friction coefficient. However, this method has obvious deficiencies in predicting the specific response of the soil when subjected to force, especially the settlement and deformation of fine-grained soil, resulting in inaccurate assessment results that are difficult to meet actual needs. Summary of the invention

[0003] The purpose of the present invention is to provide a soil particle micro-parameter calibration method, device and unmanned vehicle simulation method. By integrating a wheel loader, a soil sample platform, an environmental sensor, a machine vision system, a settlement measurement device and a numerical calculation control unit, a highly automated experimental platform is constructed. The platform can apply different loads to the soil under controlled conditions and monitor the settlement and deformation of soil particles under stress in real time. More importantly, the present invention adopts an innovative inversion algorithm to automatically adjust and optimize the micro-parameters of soil particles by comparing the differences between the simulation calculation results and the experimental measurement results until the two reach the best match. This not only can accurately determine the key parameters that affect the mechanical properties of the soil, but also provides reliable data support for subsequent soil-wheel interaction simulation.

[0004] To achieve the above object, the present invention provides the following technical solutions:

[0005] A soil particle microscopic parameter calibration method comprises the following steps:

[0006] Step 1: applying a predetermined load to a soil sample, and obtaining an actual soil settlement amplitude under the predetermined load;

[0007] Step 2, obtaining soil sample images and extracting soil particle geometric parameters;

[0008] Step 3: Input soil particle microscopic parameters;

[0009] Step 4, calculating the soil particle mesoscopic parameters through an iterative optimization algorithm according to the actual soil settlement amplitude in step 1, the soil particle geometric parameters in step 2, and the soil particle mesoscopic parameters input in step 3;

[0010] Step 5: Output the calculated soil particle mesoscopic parameters as calibrated soil mesoscopic parameters.

[0011] As a further solution of the present invention: in step 1, a predetermined load is applied to the soil sample by a wheel loader, and the settlement amplitude of the soil during the loading process is detected by a laser displacement sensor to obtain ambient temperature and ambient humidity information.

[0012] As a further solution of the present invention: in step 2, a soil image is acquired by a high-resolution camera, and geometric parameters of soil particles are acquired by image processing.

[0013] As a further solution of the present invention: the iterative optimization algorithm in step 4 adopts a back propagation algorithm based on a neural network; the iterative optimization algorithm in step 4 includes the following steps:

[0014] Step 4.1, calculating the theoretical soil settlement amplitude by using the soil particle mesoscopic parameters input in step 3, the soil particle geometric parameters extracted in step 2, and the applied predetermined load;

[0015] Step 4.2: Compare the actual soil settlement amplitude obtained in step 1 with the theoretical soil settlement amplitude obtained in step 4.1. If the difference between the theoretical soil settlement amplitude and the actual soil settlement amplitude is outside the error threshold, perform iterative calculation based on the back propagation algorithm of the neural network; if the difference between the theoretical soil settlement amplitude and the actual soil settlement amplitude is within the error threshold, the iteration ends and step 5 is executed to record and output the soil micro-particle parameters.

[0016] As a further solution of the present invention: the soil particle microscopic parameters in the arrangement 4.2 are adjusted according to a set step size.

[0017] As a further solution of the present invention: the iterative algorithm input X = [k1, k2, k3, k4, k5, k6, k7, k8, k9], the iterative algorithm output Y = [h1, h2, h3, h4], the iterative algorithm loss function L is as follows:

[0018]

[0019] The loss function L is the error between the output value of the calculated n samples and the predicted value of the simulation calculation. When the loss function L converges to a value less than the error threshold L0, the difference between the theoretical soil settlement amplitude and the actual soil settlement amplitude is within the error threshold, and the soil particle microscopic parameters at this time are output as calibration parameters.

[0020] in;

[0021] k1 represents the normal contact stiffness parameter of the particle itself, k2 represents the tangential contact stiffness parameter of the particle itself, k3 represents the particle friction coefficient, k4 represents the normal adhesion parameter of the particle itself, k5 represents the tangential adhesion parameter of the particle itself, k6 represents the average diameter of the particle, k7 represents the load applied by the wheel load meter, k8 represents the ambient temperature, and k7 represents the ambient humidity;

[0022] h1 represents the actual measured soil settlement height value at sensor position 1, h2 represents the actual measured soil settlement height value at sensor position 2, h3 represents the actual measured soil settlement height value at sensor position 3, and h4 represents the actual measured soil settlement height value at sensor position 4.

[0023] A soil particle microscopic parameter calibration device, comprising:

[0024] A soil sample table, used for placing soil samples;

[0025] A wheel loader, arranged above the soil sample platform, for applying a load to the soil sample on the soil sample platform;

[0026] A settlement measuring device for measuring the settlement amplitude of a soil sample;

[0027] A machine vision system to obtain geometric parameters of soil samples;

[0028] The numerical calculation unit is connected to the machine vision system, the wheel loader, and the settlement measurement device for receiving and processing signals.

[0029] As a further solution of the present invention: the soil sample set includes a fixed bracket and a sample arranged between the fixed bracket, and the wheel loader includes a loading wheel, a load control system, a pressure sensor and a moving mechanism, the moving mechanism is arranged at the upper end of the fixed bracket, the load control system is located below the moving mechanism and connected between the loading wheel and the moving mechanism, and the loading wheel is provided with a pressure sensor.

[0030] As a further solution of the present invention: the machine vision system includes a high-resolution camera, a light source system, and image processing software, and the settlement measurement device includes a plurality of laser displacement sensors, and the laser displacement sensors are fixedly connected to the soil sample table through a sensor mounting plate;

[0031] Also included is an environmental sensor, which includes a temperature sensor and a humidity sensor.

[0032] A method for simulating an unmanned vehicle, wherein the soil microscopic parameters outputted in step 5 are input into an unmanned vehicle simulation system, and the unmanned vehicle simulation is performed through the unmanned vehicle simulation system and the soil microscopic parameters.

[0033] Compared with the prior art, the present invention has the following beneficial effects:

[0034] 1. This application improves the accuracy of soil particle mesoscopic parameter calibration and improves calibration precision through high-resolution images and iterative optimization algorithms;

[0035] 2. The application has a high level of automation. The entire calibration process from data collection to parameter optimization can be completed automatically, which reduces manual intervention, improves work efficiency, reduces experimental costs and time, and is suitable for large-scale soil sample analysis;

[0036] 3. This application can not only be applied to the design and testing of unmanned vehicles, but can also be extended to the fields of agricultural machinery, military equipment, etc. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] Figure 1 This is a block diagram of the calibration device of this embodiment;

[0038] Figure 2 This is a flow chart of the calibration method of this embodiment;

[0039] Figure 3 This is a schematic diagram of the structure of the calibration device of this embodiment;

[0040] In the figure:

[0041] 100-wheel loader, 101-loading wheel, 102-load control system, 103-moving mechanism, 104-pressure sensor;

[0042] 110-machine vision system, 111-high-resolution camera, 112-light source system, 113-image processing software;

[0043] 120- numerical calculation control unit;

[0044] 130-soil sample table, 131-sample box, 132-fixed bracket;

[0045] 140-sedimentation measuring device, 141-laser displacement sensor, 142-sensor mounting plate;

[0046] 150 - environmental sensor, 151 - temperature sensor, 152 - humidity sensor. DETAILED DESCRIPTION

[0047] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0048] See also Figure 1 In an embodiment of the present invention, a soil particle microscopic parameter calibration device includes:

[0049] Wheel loader 100: Wheel loader 100 is a device used to simulate the force generated when an unmanned wheel contacts the ground. It consists of a loading wheel 101, a load control system 102, a pressure sensor 104, and a moving mechanism 103. The loading wheel 101 is the part that simulates the tire of a vehicle. It is a wheel that can withstand and transmit a preset load. The load control system 102 is responsible for regulating and controlling the force applied to the soil by the loading wheel 101. This system usually includes an electric or hydraulic drive device, and different load levels can be set according to experimental requirements to simulate different vehicle weights or load conditions. The pressure sensor 104 can record the force generated when the loading wheel 101 contacts the soil in real time, providing basic data for subsequent data analysis. The moving mechanism 103 enables the loading wheel 101 to move up and down on the soil surface to simulate the situation when the vehicle is driving. The moving mechanism 103 drives the loading wheel to move through a motor.

[0050] Soil sample table 130: The soil sample table 130 is an important component of the soil particle microscopic parameter calibration method and device for unmanned vehicle mobility simulation evaluation. The soil sample table 130 consists of a sample box 131 and a fixed bracket 132. The sample box 131 is a square box made of metal material, which is intended to provide a stable and controllable bearing platform for soil samples. There is a fixed bracket 132 above the sample box 131 for installing the wheel loader 100 and the settlement measurement device.

[0051] Machine vision system 110: The machine vision system 110 is mainly used in the system to capture and analyze the microscopic features of soil particles to obtain detailed information about the soil particles. The machine vision system 110 is mainly composed of a high-resolution camera 111, a light source system 112 and an image processing software 113. The high-resolution camera 111 is used to take high-definition images of soil particles to ensure that the details of the soil particles can be clearly captured. The light source system 112 provides appropriate lighting conditions to ensure the quality of the image. The light source system 112 is a ring-shaped LED lighting device to reduce the impact of shadows and reflections. The image processing software 113 includes functional modules such as image enhancement, segmentation, and feature extraction, which are used to analyze the captured image data and extract useful information therefrom, such as the size of soil particles.

[0052] Settlement measurement device: The settlement measurement device is used to accurately measure the settlement amplitude of soil during loading and is an important tool for verifying the calibration results of soil particle microscopic parameters. The device consists of four high-precision laser displacement detection sensors 141 and a sensor mounting plate 142. The four high-precision laser displacement detection sensors 141 are arranged in a rectangular row on the sensor mounting plate 142 and are used for non-contact measurement of soil displacement changes. The settlement measurement device is installed on the fixed bracket 132 above the soil sample platform 130.

[0053] Environmental sensor 150: includes a temperature sensor 152 and a humidity sensor 151, which are used to monitor and record the temperature and humidity of the test environment in real time.

[0054] The numerical calculation control unit 120 is the brain of the entire system, responsible for the movement and loading control of the wheel loader 100, processing the data collected from each sensor, and calculating the microscopic parameters of the soil particles through an iterative optimization algorithm. The calculated parameters are used for the simulation evaluation of the mobility of the unmanned vehicle to improve the accuracy and efficiency of the evaluation. The numerical calculation control unit 120 is connected to the load control system 102 of the wheel loader 100 through the CAN port to send motion control instructions. The numerical calculation control unit 120 is connected to the machine vision system 110 through the network port to obtain the size and information of the soil particles. The numerical calculation control unit 120 is connected to the environmental sensor 150 through the serial port to obtain the temperature and humidity of the experimental environment. The numerical calculation control unit 120 is connected to the settlement measurement device through the USB port to obtain the soil settlement displacement change data.

[0055] A soil particle microscopic parameter calibration method comprises the following steps:

[0056] Step 1: applying a predetermined load to a soil sample and obtaining the actual soil settlement amplitude under the predetermined load; applying a predetermined load to the soil sample by a wheel loader, detecting the settlement amplitude of the soil during the loading process by a laser displacement sensor, and obtaining ambient temperature and ambient humidity information at the same time;

[0057] Step 2: Acquire soil sample images and extract soil particle geometric parameters. Acquire soil images using a high-resolution camera and obtain soil particle geometric parameters through image processing.

[0058] Step 3: Input soil particle microscopic parameters;

[0059] Step 4, calculating the soil particle mesoscopic parameters through an iterative optimization algorithm according to the actual soil settlement amplitude in step 1, the soil particle geometric parameters in step 2, and the soil particle mesoscopic parameters input in step 3;

[0060] The iterative optimization algorithm in step 4 adopts a back propagation algorithm based on a neural network; the iterative optimization algorithm in step 4 includes the following steps:

[0061] Step 4.1, calculating the theoretical soil settlement amplitude by using the soil particle mesoscopic parameters input in step 3, the soil particle geometric parameters extracted in step 2, and the applied predetermined load;

[0062] Step 4.2: Compare the actual soil settlement amplitude obtained in step 1 with the theoretical soil settlement amplitude obtained in step 4.1. If the difference between the theoretical soil settlement amplitude and the actual soil settlement amplitude is outside the error threshold, perform iterative calculation based on the back propagation algorithm of the neural network; if the difference between the theoretical soil settlement amplitude and the actual soil settlement amplitude is within the error threshold, the iteration ends and step 5 is executed to record and output the soil micro-particle parameters.

[0063] As a further solution of the present invention: the soil particle microscopic parameters in the arrangement 4.2 are adjusted according to a set step size.

[0064] As a further solution of the present invention: the iterative algorithm input X = [k1, k2, k3, k4, k5, k6, k7, k8, k9], the iterative algorithm output Y = [h1, h2, h3, h4], the iterative algorithm loss function L is as follows:

[0065]

[0066] The loss function L is the error between the output value of the calculated n samples and the predicted value of the simulation calculation. When the loss function L converges to a value less than the error threshold L0, the difference between the theoretical soil settlement amplitude and the actual soil settlement amplitude is within the error threshold, and the soil particle microscopic parameters at this time are output as calibration parameters.

[0067] in;

[0068] k1 represents the normal contact stiffness parameter of the particle itself, k2 represents the tangential contact stiffness parameter of the particle itself, k3 represents the particle friction coefficient, k4 represents the normal adhesion parameter of the particle itself, k5 represents the tangential adhesion parameter of the particle itself, k6 represents the average diameter of the particle, k7 represents the load applied by the wheel load meter, k8 represents the ambient temperature, and k7 represents the ambient humidity;

[0069] h1 represents the actual measured soil settlement height value at sensor 1, h2 represents the actual measured soil settlement height value at sensor 2, h3 represents the actual measured soil settlement height value at sensor 3, and h4 represents the actual measured soil settlement height value at sensor 4;

[0070] Step 5: Output the calculated soil particle mesoscopic parameters as calibrated soil mesoscopic parameters.

[0071] A method for simulating an unmanned vehicle, wherein the soil microscopic parameters outputted in step 5 are input into an unmanned vehicle simulation system, and the unmanned vehicle simulation is performed through the unmanned vehicle simulation system and the soil microscopic parameters. Specific embodiments

[0073] A soil particle microscopic parameter calibration device

[0074] Wheel Loader

[0075] Loading wheel: A hard wheel with a diameter of 300 mm and a width of 150 mm, simulating the tire of an unmanned vehicle.

[0076] Load Control System: Equipped with a hydraulic drive, it can set and maintain a load range of 5kN to 20kN to simulate vehicles of different weight classes.

[0077] Pressure sensor: installed under the loading wheel, it can monitor the pressure distribution and displacement changes of the contact surface between the loading wheel and the soil in real time.

[0078] Moving mechanism: driven by a servo motor, moves along the guide rail to control the up and down movement of the loading wheel.

[0079] Soil sample stand

[0080] Size: Square with length and width of 1m and depth of 0.5m, made of stainless steel.

[0081] Function: Provide a stable testing platform for soil samples to ensure that no deviation occurs during loading.

[0082] Machine Vision System

[0083] High-resolution camera: Equipped with a 10-megapixel industrial-grade CCD camera to capture high-definition images of soil particles.

[0084] Light source system: Uses a ring LED light to provide uniform lighting, reduce shadows, and ensure image quality.

[0085] Image processing software: It has functions such as image enhancement, segmentation, and feature extraction, and can identify the size of soil particles.

[0086] Sedimentation measuring device

[0087] High-precision laser displacement detection sensor: installed above the sample table, in a rectangular array with a spacing of 20 cm, used for non-contact measurement of soil settlement displacement.

[0088] Data acquisition module: converts sensor signals into digital data and transmits them to the numerical calculation control unit via USB or RS232 interface.

[0089] Environmental Sensors

[0090] Includes temperature and humidity sensors to monitor and record experimental environmental conditions.

[0091] Numerical calculation control unit

[0092] Hardware configuration: high-performance workstation with powerful data processing capabilities.

[0093] Software system: integrates load control, data acquisition, image processing and simulation calculation functions.

[0094] A method for calibrating soil particle microscopic parameters

[0095] Step 1: Apply a predetermined load to the soil sample, set the load amplitude of the wheel loader to 3 kN to simulate the load condition of a single wheel of a medium-weight vehicle, and obtain the actual soil settlement amplitude under the predetermined load.

[0096] Step 2, obtaining soil sample images and extracting soil particle geometric parameters;

[0097] Step 3: Input the microscopic parameters of soil particles; set the initial parameters of soil particles, the normal contact stiffness parameter k1 is 1e7 N / m, the tangential contact stiffness parameter k2 is 1e6 N / m, the particle friction coefficient k3 is 0.5, and the normal adhesion parameter k4 is 100 N / m 2 , the tangential bonding strength parameter k5 is 50N / m 2 .

[0098] Step 4, calculating the soil particle mesoscopic parameters through an iterative optimization algorithm according to the actual soil settlement amplitude in step 1, the soil particle geometric parameters in step 2, and the soil particle mesoscopic parameters input in step 3;

[0099] Step 4.1, the load is set to 3KN, the average diameter of soil particles is 0.3mm, the experimentally measured settlement amplitude is 5mm, and the error threshold is set to 0.1mm. In the initial iteration, the simulation software is used to calculate the soil settlement amplitude. Compare the settlement amplitude predicted by the model with the settlement amplitude measured experimentally. If the error exceeds the set threshold, adjust the microscopic parameters and rerun.

[0100] Step 4.2, first iteration: The calculated settlement is 5.5 mm. The absolute error is 0.5 mm, which exceeds the set error threshold. Reduce the normal contact stiffness k1 to 8e6 N / m.

[0101] Second iteration: The calculated settlement is 4.8 mm. The absolute error is 0.2 mm, still above the threshold. Increase the tangential contact stiffness k2 to 1.2e6 N / m.

[0102] Third iteration: The calculated settlement amplitude is 5.1 mm. The absolute error is 0.1 mm, which meets the error threshold. Stop the iteration and record the final microscopic parameters as the calibration result.

[0103] Step 5: Output the calculated soil particle mesoscopic parameters as calibrated soil mesoscopic parameters.

[0104] Through the operation of the above specific embodiments, high-precision calibration of soil particle microscopic parameters is achieved, providing reliable data support for the mobility simulation of unmanned vehicles.

[0105] A method for simulating an unmanned vehicle

[0106] In order to evaluate the passability of unmanned vehicles under different soil conditions, the discrete element method (DEM) based on soil mesoscopic particle parameters was used for simulation calculation. First, the geometric parameters and physical properties of soil particles, such as particle size distribution, were obtained through laboratory tests and image analysis. Then, these data were used to calibrate the mesoscopic parameters in the DEM model, including normal contact stiffness, tangential contact stiffness, particle friction coefficient, normal adhesion, and tangential adhesion. Then, the dynamic model of the unmanned vehicle was established, considering factors such as the mass, center of mass position, and tire characteristics of the vehicle. By coupling the unmanned vehicle model with the soil DEM model, the interaction between the tire and the soil during the vehicle's driving process was simulated, and the soil deformation, settlement, and shear stress responses were calculated. Finally, the simulation results were analyzed to evaluate the traction performance, sinking depth, and driving stability of the unmanned vehicle, so as to optimize the design and control strategy of the unmanned vehicle and improve its passability in complex terrain.

[0107] In the design process of a new type of unmanned vehicle, this embodiment is used to calibrate the mechanical performance parameters of different soil types. The calibrated parameters are input into the simulation platform to simulate the driving conditions of the unmanned vehicle on different soft roads. The key performance indicators such as the vehicle's pass rate, stability, and energy consumption are evaluated. The design parameters of the vehicle, such as tire type and suspension system, are adjusted according to the simulation results, which effectively improves the maneuverability of the vehicle. The effectiveness and reliability of this embodiment have been proven through multiple tests and verifications.

[0108] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

[0109] In addition, it should be understood that although the present specification is described according to implementation modes, not every implementation mode contains only one independent technical solution. This description of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment may also be appropriately combined to form other implementation modes that can be understood by those skilled in the art.

Claims

1. A soil particle micro-parameter calibration method, characterized in that: The following steps are involved: Step 1: applying a predetermined load to a soil sample, and obtaining an actual soil settlement amplitude under the predetermined load; Step 2, obtaining soil sample images and extracting soil particle geometric parameters; Step 3: Input soil particle microscopic parameters; Step 4, calculating the soil particle mesoscopic parameters through an iterative optimization algorithm according to the actual soil settlement amplitude in step 1, the soil particle geometric parameters in step 2, and the soil particle mesoscopic parameters input in step 3; Step 5: Output the calculated soil particle mesoscopic parameters as calibrated soil mesoscopic parameters.

2. A soil particle microscopic parameter calibration method and device according to claim 1, characterized in that: In the step 1, a predetermined load is applied to the soil sample by a wheel loader, and the settlement amplitude of the soil during the loading process is detected by a laser displacement sensor to obtain ambient temperature and ambient humidity information.

3. A soil particle micro-parameter calibration method and device according to claim 1, characterized in that: In step 2, a soil image is acquired by a high-resolution camera, and geometric parameters of soil particles are acquired by image processing.

4. A soil particle microscopic parameter calibration method and device according to claim 1, characterized in that: The iterative optimization algorithm in step 4 adopts a back propagation algorithm based on a neural network; the iterative optimization algorithm in step 4 includes the following steps: Step 4.1, calculating the theoretical soil settlement amplitude by using the soil particle mesoscopic parameters input in step 3, the soil particle geometric parameters extracted in step 2, and the applied predetermined load; Step 4.2: Compare the actual soil settlement amplitude obtained in step 1 with the theoretical soil settlement amplitude obtained in step 4.

1. If the difference between the theoretical soil settlement amplitude and the actual soil settlement amplitude is outside the error threshold, perform iterative calculation based on the back propagation algorithm of the neural network; if the difference between the theoretical soil settlement amplitude and the actual soil settlement amplitude is within the error threshold, the iteration ends and step 5 is executed to record and output the soil micro-particle parameters.

5. A soil particle microscopic parameter calibration method and device according to claim 4, characterized in that: In the arrangement 4.2, the soil particle microscopic parameters are adjusted according to the set step size.

6. A soil particle microscopic parameter calibration method and device according to claim 4 or 5, characterized in that: The iterative algorithm input X = [k1, k2, k3, k4, k5, k6, k7, k8, k9], the iterative algorithm output Y = [h1, h2, h3, h4], the iterative algorithm loss function L is as follows: The loss function L is the error between the output value of the calculated n samples and the predicted value of the simulation calculation. When the loss function L converges to a value less than the error threshold L0, the difference between the theoretical soil settlement amplitude and the actual soil settlement amplitude is within the error threshold, and the soil particle microscopic parameters at this time are output as calibration parameters. in; k1 represents the normal contact stiffness parameter of the particle itself, k2 represents the tangential contact stiffness parameter of the particle itself, k3 represents the particle friction coefficient, k4 represents the normal adhesion parameter of the particle itself, k5 represents the tangential adhesion parameter of the particle itself, k6 represents the average diameter of the particle, k7 represents the load applied by the wheel load meter, k8 represents the ambient temperature, and k7 represents the ambient humidity; h1 represents the actual measured soil settlement height value at sensor position 1, h2 represents the actual measured soil settlement height value at sensor position 2, h3 represents the actual measured soil settlement height value at sensor position 3, and h4 represents the actual measured soil settlement height value at sensor position 4.

7. A device for calibrating a soil particle microscopic parameter according to any one of claims 1 to 6, characterized in that: include: A soil sample stand (103) for placing soil samples; A wheel loader (100) is arranged above the soil sample platform (103) and is used to apply a load to the soil sample on the soil sample platform (103); A settlement measuring device (140) for measuring the settlement amplitude of a soil sample; A machine vision system (110) for obtaining geometric parameters of a soil sample; The numerical calculation unit (120) is in signal communication with the machine vision system (110), the wheel loader (100), and the settlement measurement device (140), and is used to receive and process signals.

8. The soil particle microscopic parameter calibration device according to claim 1, characterized in that: The soil sample stand (100) comprises a fixed support (132) and a sample (131) arranged between the fixed supports. The wheel loading instrument (100) comprises a loading wheel (101), a load control system (102), a pressure sensor (104) and a moving mechanism (103). The moving mechanism (103) is arranged at the upper end of the fixed support (132). The load control system (102) is located below the moving mechanism (103) and connected between the loading wheel (101) and the moving mechanism (103). The loading wheel (101) is provided with a pressure sensor (104).

9. A soil particle microscopic parameter calibration device according to claim 1, characterized in that: The machine vision system (110) comprises a high-resolution camera (111), a light source system (112), and image processing software (113); the settlement measurement device comprises a plurality of laser displacement sensors (141), and the laser displacement sensors (141) are fixedly connected to the soil sample platform (100) via a sensor mounting plate (142); An environmental sensor (150) is also included, and the environmental sensor (150) includes a temperature sensor (152) and a humidity sensor (151).

10. An unmanned vehicle simulation method using a soil particle microscopic parameter calibration method according to any one of claims 1 to 6, characterized in that: The soil mesoscopic parameters outputted in step 5 are input into the unmanned vehicle simulation system, and the unmanned vehicle simulation is performed using the unmanned vehicle simulation system and the soil mesoscopic parameters.

Images