Grain vehicle load estimation system and method

Abstract

The invention discloses a grain vehicle weight estimation system and method, and relates to the technical field of grain vehicle weight analysis, and the system comprises a laser radar, a camera, and an embedded computer. The laser radar is used for scanning a compartment of a grain truck to obtain a vehicle point cloud model and sending the vehicle point cloud model to the embedded computer; the camera is used for shooting a carriage image and sending the carriage image to the embedded computer; according to the method, the bulk grain truck weight can be estimated in a mode of sampling and detecting the grain volume weight through the stranding cage, estimation of the grain truck weight in a static state is achieved, and early warning of a suspicious counterweight grain truck is achieved through wagon balance data comparison; the grain weight is estimated in a non-contact mode, compared with a wagon balance measurement mode, a precise wagon balance measurement tool is not needed, the grain weight can be obtained through one-time measurement, and the method is suitable for fast and low-precision grain weight measurement scenes.

Description

Technical Field

[0001] The present invention relates to the technical field of grain truck load analysis, and in particular to a grain truck load estimation system and method. Background Art

[0002] At present, most of the domestic vehicle load estimation methods collect real-time data of vehicle movement through sensors, and calculate the dynamic equations or analyze the historical data through algorithms to obtain the approximate vehicle load. The existing invention patents CN118520201A, a vehicle load estimation method based on LSTM neural network, and CN116821775A, a load estimation method based on machine learning, have relatively similar solutions, but these methods are basically for moving vehicles to estimate the load, and cannot meet the load estimation when the vehicle is stationary. When grain is purchased and stored, it is transported to each warehouse by truck and stored in the warehouse after weighing on the scale. This method has a low degree of intelligence and is easily affected by human operation. In addition, the current vehicle load estimation method is basically carried out when the vehicle is in motion, and it does not have the application conditions for estimating the load of grain trucks similar to those in a stationary state; grain depots generally use the method of taking the average of the weight of the grain truck passing through the scale twice to estimate the load of the grain truck, which is complicated and the result is easily affected by human intervention. Summary of the invention

[0003] The purpose of the present invention is to solve the problems mentioned in the above background technology and to propose a grain truck load estimation system and method.

[0004] The purpose of the present invention can be achieved through the following technical solutions:

[0005] In a first aspect, the present invention provides a grain truck load estimation system, including a ceiling rail, a camera, a laser radar, a grain sampling cage, a bulk density detection device, and an embedded computer;

[0006] The bottom end surface of the ceiling rail is installed on the ground through three columns, and a camera and a laser radar are installed on the middle column; a grain sampling cage is installed on one side of the ceiling rail, and a support platform guide rail is installed on the upper end surface of the ceiling rail, and a bulk density support platform is installed on the support platform guide rail. The bulk density support platform moves horizontally on the ceiling rail through the support platform guide rail, and a bulk density detection device is installed on the bulk density support platform; the laser radar and the camera are both communicatively connected to the embedded computer.

[0007] As a preferred embodiment of the present invention, the installation positions of the camera and the laser radar are higher than the upper end surface of the vehicle body, and the laser radar is located above the camera.

[0008] As a preferred embodiment of the present invention, the grain sampling cage includes a cage shell, an electric hydraulic support rod, a rotating blade, a discharge port and a motor; the middle part of the cage shell is installed on the ceiling rail through two electric hydraulic support rods, an intermediate shaft and rotating blades installed on the intermediate shaft are installed inside the cage shell, a motor is installed at one end of the cage shell, a discharge port is opened at one end of the cage shell close to the motor, and the output shaft of the motor is transmission connected to the intermediate shaft.

[0009] As a preferred embodiment of the present invention, the bulk density detection device includes a supporting frame, a star-shaped discharge valve is installed on the supporting frame, a feed port is installed at the upper end of the star-shaped discharge valve, a proximity sensor is installed on one side of the top of the feed port, and the installation position of the proximity sensor is higher than the upper end surface of the feed port; a pneumatic scraping mechanism is installed in the middle of the interior of the supporting frame, a receiving cup is installed just below the star-shaped discharge valve, the receiving cup is placed on a weighing scale, and the weighing scale and the supporting frame are fixedly installed on a bulk density support table; a discard hopper is installed on one side of the receiving cup; and the weighing scale is communicatively connected to the embedded computer.

[0010] As a preferred embodiment of the present invention, the pneumatic scraper mechanism includes two parallel slide rails fixedly installed inside the support frame, slide rail cylinders are installed on the two slide rails, a fixed plate is installed between the two slide rail cylinders, and the two sides of the fixed plate are respectively fixed on the two slide rail cylinders, and a scraper tube is installed in the middle of the fixed plate.

[0011] In a second aspect, the present invention provides a method for estimating the weight of a grain truck, comprising:

[0012] The laser radar scans the compartment of the grain truck to obtain a vehicle point cloud model and sends it to the embedded computer; the camera takes an image of the compartment and sends it to the embedded computer;

[0013] Calibrate the camera and lidar to obtain the camera-to-lidar transformation relationship, which is used to map the vehicle floor pixels to the radar point cloud;

[0014] The vehicle point cloud model is automatically identified and segmented to obtain a pixel-level point cloud model of the vehicle compartment. The four corner points of the compartment are extracted, and then the vehicle 3D data is constructed. The four corner points of the grain truck are located, and the 3D point cloud of the grain in the compartment is cut out. The fused data is used to extract the compartment floor to obtain the reference surface for measuring the volume of the grain on board.

[0015] By dividing the grain 3D point cloud into triangular meshes, the base plane of the carriage floor is obtained. Based on the triangular mesh processing technology, mesh smoothing, defect repair, and filling processing are performed. The base plane of the pile is specified, and the volume microelement between each triangle and the base plane is calculated. The volume of all volume microelement is summed up to obtain the total volume of the grain on the vehicle.

[0016] The weighing balance will measure the weight of the corresponding grain per unit volume in the receiving cup and send it to the embedded computer. After receiving the weight of the grain per unit volume, the embedded computer will convert the volume to get the grain load of the whole vehicle and compare it with the data of the grain truck through the scale. If the weight difference between the two is within the preset range, the grain loading in the granary is normal. If the weight difference exceeds the preset range, a suspicious weight distribution instruction is generated and a corresponding warning is issued.

[0017] As a preferred embodiment of the present invention, the specific process of the embedded computer converting the received unit volume weight of grain into volume to obtain the whole vehicle grain load is as follows: mark the unit volume weight of grain as TM, the total volume of grain on the vehicle as ZT, and the unit volume as MS, and substitute into the formula ZG=ζ×TM×(ZT / MS) to obtain the whole vehicle grain load ZG; wherein ζ is the grain repair factor.

[0018] As a preferred embodiment of the present invention, the specific calculation process of the grain modification factor is as follows: randomly collecting a number of grain samples in a grain cart and a number of grain samples in a receiving cup, photographing the grain samples to obtain sample images corresponding to the grain samples, identifying the contours of grain particles in the sample images to obtain particle contours, identifying the area corresponding to the particle contours and extracting their numerical values ​​to obtain profile values; summing up the profile values ​​of all grain particles in the sample images and taking the average to obtain a profile mean; weighing the grain samples to obtain sample weights, summing up the profile means of all grain samples corresponding to the grain cart to obtain a profile total value one, and calculating the average value of all sample weights corresponding to the grain cart to obtain a weight mean one; summing up the profile means of all grain samples corresponding to the receiving cup to obtain a profile total value two, and calculating the average value of all sample weights corresponding to the grain cart to obtain a weight mean two; normalizing the profile total value one, the weight mean one, the profile total value two and the weight mean two, taking the numerical values ​​of the four, and substituting them into the calculation model to output the grain modification factor ζ.

[0019] As a preferred implementation of the present invention, the calculation model is specifically: Among them, KM1 is the value of the profile total value one, KM2 is the value of the profile total value two, TS1 is the value of the weight mean one, TS2 is the value of the weight mean two, ya1 is the profile threshold, Ta1 is the weight threshold, λ1, λ2 and λ3 are preset weight factors.

[0020] Compared with the prior art, the present invention has the following beneficial effects:

[0021] 1. The present invention can estimate the load of bulk grain trucks by testing the bulk density of grain through cage sampling, thereby realizing the estimation of the load of grain trucks in a stationary state, and realizing early warning of suspicious counterweight grain trucks through weighbridge data comparison.

[0022] 2. The present invention estimates the weight of grain in a contactless manner. Compared with the floor scale measurement method, it does not require precise floor scale measurement tools and can obtain the weight of grain in one measurement. It is suitable for fast grain weight measurement scenarios with low precision requirements. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0024] Figure 1 It is a principle block diagram of the present invention;

[0025] Figure 2 It is a schematic diagram of the structure of the ceiling rail, grain sampling cage and bulk density detection device of the present invention;

[0026] Figure 3 This is a schematic diagram of the overall structure of the grain sampling cage of the present invention;

[0027] Figure 4 It is a schematic diagram of the overall structure of the bulk density detection device of the present invention;

[0028] Figure 5 It is a schematic diagram of the overall structure of the pneumatic scraping mechanism of the present invention;

[0029] Figure 6 It is a schematic diagram of calculating the volume of grain on board the vehicle according to the present invention;

[0030] Figure 7 This is a principle block diagram of an embedded computer according to Embodiment 2 of the present invention.

[0031] Reference numerals:

[0032] 1. Overhead rail; 2. Column; 3. Camera; 4. LiDAR; 5. Grain sampling cage; 6. Bulk weight support platform; 7. Bulk weight detection device; 51. Cage shell; 52. Electric hydraulic support rod; 53. Rotating blade; 54. Discharge port; 55. Motor; 71. Support frame; 72. Inlet; 73. Star-shaped discharge valve; 74. Receiving cup; 75. Pneumatic scraper mechanism; 76. Weighing scale; 77. Discard hopper; 78. Proximity sensor; 751. Slide rail; 752. Slide rail cylinder; 753. Fixed plate; 754. Scraper tube. DETAILED DESCRIPTION

[0033] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than 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.

[0034] Example 1

[0035] See also Figure 1-2 As shown, a grain truck load estimation system includes a ceiling rail 1, a grain sampling cage 5, a bulk density detection device 7 and an embedded computer; the bottom end surface of the ceiling rail 1 is installed on the ground through three columns 2, wherein a camera 3 and a laser radar 4 are installed on the middle column 2, the installation positions of the camera 3 and the laser radar 4 are higher than the upper end surface of the car body, and the laser radar 4 is located above the camera 3; the laser radar 4 is used to scan the car body of the grain truck to obtain a vehicle point cloud model and send it to the embedded computer; the camera 3 is used to take a car body image and send it to the embedded computer; the embedded computer is used to receive the vehicle point cloud model and the car body image and analyze them to obtain the total volume of the grain on the vehicle, and the specific analysis process is: calibrate the camera and the laser radar, obtain the camera to laser radar transformation relationship, and use it to convert the vehicle bottom plate pixel The points are mapped to the radar point cloud; the vehicle point cloud model is automatically identified and segmented to obtain a pixel-level car point cloud model, and the four corner points of the car are extracted. Then, the vehicle three-dimensional data is constructed, and the four corner points of the grain car are located (this is the prior art, specifically refer to CN116148809A for the automatic generation of grain car sampling points based on laser radar scanning and positioning and the system disclosed in the process of extracting the coordinates of the corner points of the grain car), the 3D point cloud of the grain in the car is cut out, and the car floor is extracted by using the fused data to obtain the reference surface for measuring the volume of the on-board grain. The car floor reference surface is obtained by dividing the grain 3D point cloud triangular mesh, and the mesh smoothing, defect repair, filling and other processing are performed based on the triangular mesh processing technology. The reference surface of the pile is specified, and the volume microelement between each triangle and the reference surface is calculated (such as Figure 6 As shown in the figure, the volumes of all volume elements are summed up to obtain the total volume of food on board.

[0036] A grain sampling cage 5 is installed on one side of the ceiling rail 1, a support platform guide rail is installed on the upper end surface of the ceiling rail 1, a bulk density support platform 6 is installed on the support platform guide rail, and the bulk density support platform 6 moves horizontally on the ceiling rail 1 through the support platform guide rail, and a bulk density detection device 7 is installed on the bulk density support platform 6.

[0037] See also Figure 2-3As shown, the grain sampling cage 5 is used to be inserted into the grain pile of the grain truck and take samples, and then the sampled grain is transported to the bulk density detection device 7, and comprises a cage shell 51, an electric hydraulic support rod 52, a rotating blade 53, a discharge port 54 and a motor 55; the middle part of the cage shell 51 is installed on the overhead rail 1 through two electric hydraulic support rods 52, an intermediate shaft and a rotating blade 53 installed on the intermediate shaft are installed inside the cage shell 51, a motor 55 is installed at one end of the cage shell 51, and a discharge port 54 is opened at one end of the cage shell 51 close to the motor 55, the output shaft of the motor 55 is transmission-connected with the intermediate shaft, and the intermediate shaft is driven to rotate by the motor 55, thereby driving the rotating blade 53 to rotate. During use, the cage shell 51 is inserted into the grain pile of the grain truck through the electric hydraulic support rod 52, and then the rotating blade 53 rotates to transport the grain in the grain pile to the discharge port 54 and send it into the bulk density detection device 7.

[0038] See also Figure 4 As shown, the bulk density detection device 7 includes a support frame 71, on which a star-shaped discharge valve 73 is installed, and a feed port 72 is installed at the upper end of the star-shaped discharge valve 73, and a proximity sensor 78 is installed on one side of the top of the feed port 72, and the installation position of the proximity sensor 78 is slightly higher than the upper end surface of the feed port 72; a pneumatic scraper mechanism 75 is installed in the middle of the interior of the support frame 71, and a receiving cup 74 is installed just below the star-shaped discharge valve 73, and the receiving cup 74 is placed on a weighing scale 76, and the weighing scale 76 and the support frame 71 are fixedly installed on the bulk density support platform 6; a discard hopper 77 is installed on one side of the receiving cup 74; the pneumatic scraper mechanism 75 is used to scrape the grain that is higher than the horizontal plane of the corresponding cup mouth of the receiving cup 74 into the discard hopper 77 at the rear, and then discharge it through the bottom of the discard hopper inlet 77; the remaining grain in the receiving cup 74 can be weighed by the weighing scale 76 to obtain the corresponding weight, and then the grain bulk density can be obtained.

[0039] See also Figure 5 As shown, the pneumatic scraper mechanism 75 includes two parallel slide rails 751 fixedly mounted inside the support frame 71, and slide rail cylinders 752 are mounted on the two slide rails 751. A fixed plate 753 is mounted between the two slide rail cylinders 752, and the two sides of the fixed plate 753 are respectively fixed on the two slide rail cylinders 752, and a scraper tube 754 is mounted in the middle of the fixed plate 753. The slide rail cylinder 752 slides on the slide rail 751, thereby driving the fixed plate 753 and the scraper tube 754 to move, and then the food on the horizontal surface of the corresponding cup mouth of the receiving cup 74 is scraped into the rear abandoned hopper 77 through the bottom end surface of the scraper tube 754 for discharge.

[0040] The weighing scale 76 is connected to the embedded computer for communication, and sends the weight of the grain per unit volume corresponding to the grain in the measuring receiving cup 74 to the embedded computer. After receiving the weight of the grain per unit volume, the embedded computer converts the volume to obtain the grain load of the entire vehicle, and compares it with the data of the grain truck through the scale. If the weight difference between the two is within the preset range, the grain loading in the granary is normal. If the weight difference exceeds the preset range, a suspicious weight balancing instruction is generated, and a corresponding warning is issued.

[0041] Example 2

[0042] On the basis of Example 1, the specific process of the embedded computer converting the volume after receiving the unit volume weight of grain to obtain the whole vehicle grain load is as follows: mark the unit volume weight of grain as TM, mark the total volume of grain on the vehicle as ZT, mark the unit volume as MS, substitute into the formula ZG=ζ×TM×(ZT / MS) to obtain the whole vehicle grain load ZG; wherein ζ is the grain repair factor; the specific calculation process of the grain repair factor is as follows: randomly collect a number of grain samples in the grain vehicle and a number of grain samples in the receiving cup 74, and the number of grain samples in the two is the same; capture the grain sample to obtain the sample image corresponding to the grain sample, perform contour recognition on the grain particles in the sample image to obtain the particle contour, Identify the area corresponding to the particle outline and extract its value to obtain the profile value; sum the profile values ​​of all grain particles in the sample image and take the average to obtain the profile mean; weigh the grain sample to obtain the sample weight, sum the profile means of all grain samples corresponding to the grain cart to obtain the profile total value one, and calculate the average weight of all sample weights corresponding to the grain cart to obtain the weight mean one; sum the profile means of all grain samples corresponding to the receiving cup 74 to obtain the profile total value two, and calculate the average weight of all sample weights corresponding to the grain cart to obtain the weight mean two; normalize the profile total value one, weight mean one, profile total value two and weight mean two, take the values ​​of the four, and substitute them into the calculation model Output grain repair factor ζ, where KM1 is the value of the profile total value one, KM2 is the value of the profile total value two, TS1 is the value of the weight mean one, TS2 is the value of the weight mean two, ya1 is the profile threshold, Ta1 is the weight threshold, λ1, λ2 and λ3 are preset weight factors, and their sizes are custom settings, such as λ1=1.17, λ2=0.6, λ3=0.4.

[0043] By randomly collecting several grain samples in the grain truck and several grain samples in the receiving cup and analyzing them, the total profile value one, the weight mean one, the total profile value two and the weight mean two are obtained, and the grain correction factor is calculated based on this. The weight of the whole truckload of grain is corrected by the grain correction factor to make it closer to the true value, reduce the error of the sampling sample, and avoid the deviation between the sample of the grain sampling cage 5 and the whole truckload of grain, which will cause inaccurate estimation of the grain truck load and affect subsequent evaluation.

[0044] Embodiment 3:

[0045] See also Figure 7 As shown, on the basis of Example 2, a data acquisition unit and a data analysis unit are further provided in the embedded computer, wherein the data acquisition unit collects the operating information of the laser radar 4 and sends it to the data analysis unit; wherein the operating information of the laser radar 4 includes signal strength data, response delay time data, operating temperature data, operating voltage data and ambient humidity data; wherein the response delay time data includes several response delay times, which are calculated by the time difference between the moment when the laser radar 4 receives the instruction and the moment when it responds to execute the instruction.

[0046] The data analysis unit analyzes the operation information. The specific analysis process is: the signal strength data is parsed to obtain a number of signal strength values, and the signal strength value is compared with the preset signal strength range. If the signal strength value is weaker than the minimum value of the preset signal strength range, the signal strength value is marked as a weakness value; if the signal strength value is stronger than the maximum value of the preset signal strength range, the signal strength value is marked as a strength value. For example, in the preset signal strength range (A1, A2), A1 is -100dBm, and A2 is -30dBm; if the signal strength value is -110dBm, then -110dBm < -100dBm, and -110dBm is marked as a weakness value. The total number of weak values ​​is counted and marked as WD1. When the total number of weak values ​​is greater than or equal to the set number threshold WDY, the weak values ​​are sorted according to the order of acquisition time, and the time difference between two adjacent weak values ​​is calculated to obtain the signal weakness duration. The average value of all the signal weakness durations is calculated to obtain the signal weakness mean and marked as WDJ; substitute it into the preset signal analysis model Output the signal weakness base value ρf, where WDZ is the sum of the values ​​of all signal weakness values ​​without the negative sign and unit, MJZ is the sum of the values ​​of all strength values ​​without the negative sign and unit; MJmin is the value of the smallest signal weakness value among all signal weakness values ​​without the negative sign and unit; η1, η2, η3 and η4 are all preset weight coefficients, and their sizes are user-defined, and η4>η3>η1>η2>0. If the signal weakness base value is greater than or equal to the set base value threshold, a signal strength abnormality instruction of the laser radar 4 is generated and sent to the corresponding user's smart terminal to check and maintain the signal strength of the laser radar 4. If the signal weak base value is less than the set base value threshold, the response delay time data is analyzed, and several response delay times are compared with the set time threshold. If the response delay time is greater than the set time threshold, the response delay time is marked as an abnormal response time, and the average value of all abnormal response times is calculated and the value is taken to obtain the average abnormal value; the value of the largest response delay time is marked as the maximum abnormal value; the number of all abnormal response times is counted to obtain the total number of abnormalities; the average abnormal value, the total number of abnormalities and the maximum abnormal value are converted to the corresponding length according to the preset ratio, and the average abnormal value, the total number of abnormalities and the maximum abnormal value are used as ... converted to the corresponding length according to the preset ratio. Construct a triangle with the length of the total number of anomalies as the two right-angled sides, select the midpoint of the hypotenuse of the triangle, and draw a straight line with a length equal to the corresponding length of the maximum anomaly value from the midpoint to the outside of the triangle and perpendicular to the hypotenuse of the triangle. Connect the end points of the straight line to the two end points of the hypotenuse of the triangle to form a quadrilateral, calculate the area of ​​the quadrilateral and extract the numerical value of the area, and mark it as the abnormal base value DS. If the abnormal base value is greater than the set abnormal threshold, a response abnormality instruction is generated and sent to the intelligent terminal of the corresponding user to perform operation response inspection and repair on the laser radar 4. If the base value of the abnormal sound is less than or equal to the set abnormal sound threshold, the operating temperature data, the operating voltage data and the ambient humidity data are analyzed, and the operating temperature data, the operating voltage data and the ambient humidity data are compared with the corresponding temperature range, voltage range and humidity range, and the number of operating temperatures, operating voltages and ambient humidity that are not in the corresponding temperature range, voltage range and humidity range are counted respectively, and their numbers are marked as temperature outliers, pressure outliers and humidity outliers respectively, represented by the symbol SLj, j=1, 2, 3; the time of the last maintenance inspection of the laser radar 4 is obtained, if not, the time it was put into use is obtained, and the time difference between the obtained time and the current time is calculated to obtain the operating time of the laser radar 4, and substituted into the usage analysis model Output the maintenance value SW, where ST1 is the value of the operating time, ms1, ms2 and ms3 are the preset weight factors corresponding to the temperature external number, pressure external number and humidity external number, respectively, γ1 and γ2 are the preset weight ratios, and γ1+γ2=1. The sizes of the preset weight factors and the preset weight ratios are custom settings. When the maintenance value is greater than the set maintenance threshold, the maintenance value, the sound difference base value and the signal weak base value are normalized and the values ​​of the three are taken. The weights corresponding to the maintenance value, the sound difference base value and the signal weak base value are set to Ψ1, Ψ2 and Ψ3 respectively, and the maintenance value WHz is obtained using the formula WHz=SW×Ψ1+DS×Ψ2+ρf×Ψ3. When the maintenance value is greater than the maintenance threshold, the corresponding radar maintenance instruction is generated and sent to the corresponding user's smart terminal for display and warning, and the user maintains or replaces the corresponding laser radar 4. The present invention can accurately identify abnormal signal strength by comparing and analyzing the signal strength value with the preset range, which helps to more accurately judge the performance status of the laser radar 4 and avoid signal abnormality affecting the use of the laser radar 4. By analyzing the operating temperature data, operating voltage data and ambient humidity data, potential faults and problems can be discovered in time, avoiding shutdown or damage caused by laser radar 4 failure, and extending the service life of the laser radar. It provides strong support for the maintenance and management of the laser radar 4.

[0047] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A grain truck load estimation system, comprising a ceiling rail (1), a camera (3), a laser radar (4), a grain sampling cage (5), a bulk density detection device (7) and an embedded computer; characterized in that: The bottom end surface of the ceiling rail (1) is installed on the ground through three columns (2), wherein a camera (3) and a laser radar (4) are installed on the column (2) located in the middle; a grain sampling cage (5) is installed on one side of the ceiling rail (1), a support platform guide rail is installed on the upper end surface of the ceiling rail (1), a bulk density support platform (6) is installed on the support platform guide rail, and the bulk density support platform (6) moves horizontally on the ceiling rail (1) through the support platform guide rail, and a bulk density detection device (7) is installed on the bulk density support platform (6); the laser radar (4) and the camera (3) are both connected to the embedded computer for communication.

2. A grain truck load estimation system according to claim 1, characterized in that: The camera (3) and the laser radar (4) are installed at a position higher than the upper end surface of the carriage, and the laser radar (4) is located above the camera (3).

3. A grain truck load estimation system according to claim 1, characterized in that: The grain sampling cage (5) comprises a cage housing (51), an electric hydraulic support rod (52), a rotating blade (53), a discharge port (54) and a motor (55); the middle part of the cage housing (51) is mounted on the ceiling rail (1) via two electric hydraulic support rods (52); an intermediate shaft and a rotating blade (53) mounted on the intermediate shaft are mounted inside the cage housing (51); a motor (55) is mounted at one end of the cage housing (51); a discharge port (54) is provided at one end of the cage housing (51) close to the motor (55); and an output shaft of the motor (55) is drivingly connected to the intermediate shaft.

4. A grain truck load estimation system according to claim 1, characterized in that: The bulk density detection device (7) comprises a support frame (71), a star-shaped discharge valve (73) is installed on the support frame (71), a feed port (72) is installed at the upper end of the star-shaped discharge valve (73), a proximity sensor (78) is installed on one side of the top end of the feed port (72), and the installation position of the proximity sensor (78) is higher than the upper end surface of the feed port (72); a pneumatic scraper mechanism (75) is installed in the middle of the support frame (71), a receiving cup (74) is installed just below the star-shaped discharge valve (73), the receiving cup (74) is placed on a weighing scale (76), and the weighing scale (76) and the support frame (71) are fixedly installed on the bulk density support platform (6); a discard hopper (77) is installed on one side of the receiving cup (74); and the weighing scale (76) is communicatively connected with the embedded computer.

5. A grain truck load estimation system according to claim 4, characterized in that: The pneumatic scraper mechanism (75) comprises two parallel slide rails (751) fixedly mounted inside the support frame (71), the two slide rails (751) are both mounted with slide rail cylinders (752), a fixing plate (753) is mounted between the two slide rail cylinders (752), and the two sides of the fixing plate (753) are respectively fixed on the two slide rail cylinders (752), and a scraper pipe (754) is mounted in the middle of the fixing plate (753).

6. A method for estimating the weight of a grain truck, characterized in that A grain truck load estimation system applied to any one of claims 1 to 5, the method comprising: The laser radar (4) scans the compartment of the grain truck to obtain a vehicle point cloud model and sends it to the embedded computer; the camera (3) takes an image of the compartment and sends it to the embedded computer; The camera (3) and the laser radar (4) are calibrated to obtain the transformation relationship from the camera (3) to the laser radar (4), which is used to map the pixel points of the vehicle floor to the radar point cloud; the vehicle point cloud model is automatically identified and segmented to obtain a pixel-level vehicle floor point cloud model, and the four corner points of the vehicle are extracted, and then the vehicle three-dimensional data is constructed, and the four corner points of the grain truck are located, and the 3D point cloud of the grain in the vehicle is cut out, and the vehicle floor is extracted using the fused data to obtain a reference surface for measuring the volume of the grain on the vehicle; the reference surface of the vehicle floor is obtained by dividing the 3D point cloud of the grain into triangular meshes, and mesh smoothing, defect repair, and filling processing are performed based on the triangular mesh processing technology, and the reference surface of the stockpile is specified, and the volume microelement between each triangle and the reference surface is calculated, and the volume of all volume microelement is summed to obtain the total volume of the grain on the vehicle; The weighing balance (76) measures the weight of the grain per unit volume in the receiving cup and sends it to the embedded computer. After receiving the weight of the grain per unit volume, the embedded computer converts the volume to obtain the grain load of the entire vehicle and compares it with the data of the grain truck through the scale. If the weight difference between the two is within the preset range, the grain load in the granary is normal. If the weight difference exceeds the preset range, a suspicious weight distribution instruction is generated and a corresponding warning is issued.

7. A method for estimating the weight of a grain truck according to claim 6, characterized in that: After receiving the unit volume weight of grain, the embedded computer performs volume conversion to obtain the whole vehicle grain load. The specific process is: mark the unit volume weight of grain as TM, the total volume of grain on the vehicle as ZT, and the unit volume as MS, and substitute into the formula ZG=ζ×TM×(ZT / MS) to obtain the whole vehicle grain load ZG; wherein ζ is the grain repair factor.

8. A method for estimating the weight of a grain truck according to claim 7, characterized in that: The specific calculation process of the grain repair factor is as follows: randomly collecting a number of grain samples in the grain truck and a number of grain samples in the receiving cup (74), photographing the grain samples to obtain sample images corresponding to the grain samples, identifying the contours of grain particles in the sample images to obtain particle contours, identifying the area corresponding to the particle contours and extracting the area value to obtain the profile value; summing up the profile values ​​of all grain particles in the sample images and taking the average to obtain the profile mean value; The grain samples are weighed to obtain the sample weight, the profile means of all grain samples corresponding to the grain cart are summed to obtain the profile total value one, and the average weight of all sample weights corresponding to the grain cart is calculated to obtain the weight mean one; the profile means of all grain samples corresponding to the receiving cup (74) are summed to obtain the profile total value two, and the average weight of all sample weights corresponding to the grain cart is calculated to obtain the weight mean two; the profile total value one, the weight mean one, the profile total value two and the weight mean two are normalized and the values ​​of the four are taken and substituted into the calculation model to output the grain repair factor ζ.

9. A method for estimating the weight of a grain truck according to claim 8, characterized in that: The calculation model is specifically: Among them, KM1 is the value of the profile total value one, KM2 is the value of the profile total value two, TS1 is the value of the weight mean one, TS2 is the value of the weight mean two, ya1 is the profile threshold, Ta1 is the weight threshold, λ1, λ2 and λ3 are preset weight factors.

Images