Unmanned automatic stacking and transferring method and system for bulk materials

Abstract

The invention relates to the technical field of material transfer, and discloses an unmanned automatic stacking and transferring method and system for bulk materials, and the method comprises the steps: collecting the three-dimensional point cloud data of a bulk material storage yard, carrying out the splicing of the three-dimensional point cloud data, obtaining the overall point cloud of the storage yard, and building an elevation map according to the overall point cloud of the storage yard; collecting body attitude data of the bulk cargo stacking transfer truck, planning a material taking motion path on the basis of the elevation map according to the body attitude data, updating the three-dimensional point cloud data in real time, and correcting the material taking motion path; and communicating with rear-end transfer equipment, performing cooperative control on a material taking motion path, dynamically calculating the quantity of remaining materials in a storage yard according to the three-dimensional point cloud data updated in real time, and completing unmanned automatic stacking, taking and transferring of the bulk materials. Through unmanned and intelligent control, efficient, safe and stable operation of the bucket-wheel stacker-reclaimer is achieved, manual intervention is reduced, the equipment operation efficiency is improved, and the intelligent operation and maintenance requirements of a large bulk material storage yard are met.

Description

Technical Field

[0001] The present invention relates to the technical field of material transfer, and in particular to a method and system for unmanned automatic stacking and transfer of bulk materials. Background Art

[0002] With the continuous expansion of industrial production and logistics, the use and circulation of bulk materials in industries such as mining, ports, power, and coal are becoming increasingly frequent, and the demand for efficient and unmanned stacking and reclaiming in storage yards is also increasing. While traditional bucket-wheel stackers offer the advantages of continuous conveying and high production capacity, they have limited adaptability to stack heights and material hardness in dynamic stockyard environments. Furthermore, they rely on manual operation for a long time, making it difficult to strike a balance between efficiency and safety.

[0003] To achieve intelligent and unmanned operations in bulk material storage yards, existing technologies are gradually incorporating sensor monitoring, automatic control, and 3D environmental perception, aiming to achieve automated movement, precise pitching, and adaptive cutting of material surfaces. However, due to the variable distribution of materials in storage yards, the complexity of equipment posture control, and the incomplete linkage between conveying and fault warning, efficiently completing core tasks such as bulk material stacking, transfer, positioning, and obstacle avoidance remains a key challenge that the industry urgently needs to address. Summary of the Invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the technical problem solved by the present invention is: how to efficiently complete the core tasks of automatic stacking and transportation of bulk materials.

[0006] To solve the above technical problems, the present invention provides the following technical solution: a method for unmanned automatic stacking and transporting of bulk materials, comprising:

[0007] Collect 3D point cloud data of the bulk material yard, obtain the overall point cloud of the yard based on the 3D point cloud data, and create an elevation map based on the overall point cloud of the yard;

[0008] Collect the posture data of the bulk material stacking transfer vehicle, plan the material reclaiming movement path based on the elevation map based on the posture data, update the 3D point cloud data in real time and correct the material reclaiming movement path;

[0009] It communicates with the back-end transfer equipment, coordinates the material movement path, dynamically calculates the remaining material quantity in the yard based on the real-time updated 3D point cloud data, and completes unmanned automatic stacking and transfer of bulk materials.

[0010] As a preferred embodiment of the method for unmanned automatic stacking and transfer of bulk materials described in the present invention, the step of obtaining the overall point cloud of the storage yard based on the three-dimensional point cloud data includes filtering the three-dimensional point cloud data, performing preliminary coordinate transformation on each frame of the filtered three-dimensional point cloud data, and aligning the data to the same world coordinate system.

[0011] Calculate the feature descriptors of 3D point cloud data, use the RANSAC-based feature matching algorithm for coarse registration, and fine-tune the coarse registration results through fine registration to accurately align the overlapping areas;

[0012] Repeat the coarse and fine registration processes to align and merge multiple frames of 3D point cloud data into the same world coordinate system to obtain the overall point cloud of the storage yard.

[0013] As a preferred embodiment of the method for unmanned automatic stacking and transfer of bulk materials described in the present invention, the method of establishing an elevation map based on the overall point cloud of the storage yard includes setting a grid resolution on the plane (x, y) based on the overall point cloud of the storage yard using (x, y) as plane coordinates, dividing the grid cells according to the resolution, recording the maximum height of each grid cell, and obtaining an equally spaced grid elevation map.

[0014] Obtain historical data on bulk material transfers at a bulk material storage yard and assign a set of key attributes to each material, including material hardness grade, bulk density, and safe slope threshold.

[0015] Obtain the key attribute range of each material based on the historical data of the material and establish a material database;

[0016] When the staff creates a stacking and transfer task, the key attributes of the corresponding material type in the material database are called according to the material type entered by the operator, and the elevation map is layered.

[0017] As a preferred solution of the method for unmanned automatic stacking and transporting of bulk materials described in the present invention, wherein: the stratification of the elevation map includes obtaining the overall height interval of the yard according to the elevation map [z min , z max ], generate benchmark layer thickness based on key material attributes;

[0018] The slope of each grid cell is read from the elevation map. For each grid cell, the local layer thickness is calculated based on the slope and the hardness of the material, expressed as:

[0019] Δh local (i) = Δh base -α s ·f s (S(i))-α m ·f m (Hm )

[0020] Where Δh local (i) represents the local layer thickness of grid cell i; Δh base represents the base layer thickness; α s Indicates the slope correction coefficient; f s (S(i)) represents the slope mapping function; S(i) represents the slope of grid cell i; α m Indicates the hardness correction factor; f m (H m ) represents the hardness mapping function; H m Indicates the hardness of material m;

[0021] The slope and hardness are mapped to 0 or positive values through the slope mapping function and the hardness mapping function, and the quantification of Δh base The reduction strength of the calculated Δh local (i) Perform clipping to obtain the actual layer thickness for each region.

[0022] As a preferred embodiment of the method for unmanned automatic stacking and transfer of bulk materials described in the present invention, the method includes planning the material movement path based on the elevation map, obtaining the ground height H(i, j) corresponding to each grid cell (i, j) according to the elevation map, calculating the slope difference between adjacent grid cells, and marking the grid cell (i, j) as infeasible if the calculated slope difference Slope(i, j) exceeds a safe slope threshold;

[0023] The fuselage posture data includes the fuselage heading angle and bucket wheel inclination angle. The bucket wheel machine motion planning is defined on the elevation map as follows: K = {(x, y, θ, φ)}, where (x, y) represents the discrete coordinates of the fuselage on the grid, θ represents the fuselage heading angle, and φ represents the bucket wheel inclination angle.

[0024] The sampling space is constructed by the grid method. The starting position of the bulk material stacking transfer vehicle {(x0, y0, θ0, φ0)} is obtained according to the fuselage posture data. The search is performed, the neighborhood nodes are expanded in sequence, and the cost is accumulated until the material collection position is reached.

[0025] The output material movement path is a node sequence, {(x0, y0, θ0, φ0), (x1, y1, θ1, φ1), ..., (x n ,y n ,θ n ,φ n )}, indicating the trajectory of the fuselage, the change of its orientation and the change of the bucket wheel inclination angle.

[0026] As a preferred embodiment of the unmanned automatic bulk material stacking and transfer method of the present invention, the correction of the material reclaiming movement path includes acquiring three-dimensional point cloud data of the material reclaiming position in real time through a sensor during bucket wheel operation, and coordinating the data with the real-time collected fuselage posture data to locally update the elevation map;

[0027] Retrieve the height distribution H recorded in the previous cycle of the material collection location on the elevation map old (i, j), and the new height distribution H after local update new (i, j) to calculate the difference and obtain the height difference ΔH(i, j). The material removal situation is determined based on the height difference and the actual layer thickness:

[0028] If ΔH(i, j) is greater than the actual layer thickness, it is judged as excessive material extraction;

[0029] If ΔH(i, j) is less than the actual layer thickness, it is judged that insufficient material is taken;

[0030] The actual layer thickness of the next cycle is adjusted according to the material taking situation.

[0031] As a preferred solution of the unmanned automatic stacking and transferring method of bulk materials described in the present invention, the collaborative control includes obtaining the transfer rate and material flow of the rear-end transfer equipment, calculating the material taking rate based on the real-time collected fuselage posture data, and performing collaborative control based on the comparison result of the transfer rate and the material taking rate.

[0032] A bulk material unmanned automatic stacking and transfer system using any of the methods described in the present invention, wherein: a scene module collects three-dimensional point cloud data of a bulk material yard, obtains an overall point cloud of the yard based on the three-dimensional point cloud data, and establishes an elevation map based on the overall point cloud of the yard;

[0033] The path module collects the posture data of the bulk material stacking transfer vehicle, plans the material reclaiming movement path based on the elevation map based on the body posture data, updates the 3D point cloud data in real time, and corrects the material reclaiming movement path;

[0034] The collaborative module communicates with the back-end transfer equipment to collaboratively control the material retrieving movement path, dynamically calculates the amount of remaining materials in the yard based on the real-time updated three-dimensional point cloud data, and completes unmanned automatic stacking and transfer of bulk materials.

[0035] A computer device comprises: a memory and a processor; the memory stores a computer program, comprising: the steps of implementing any one of the methods of the present invention when the processor executes the computer program.

[0036] A computer-readable storage medium stores a computer program thereon, comprising: steps of implementing any one of the methods of the present invention when the computer program is executed by a processor.

[0037] Beneficial effects of the present invention: The method of the present invention realizes the acquisition of three-dimensional point cloud data by arranging multiple sensors on the bucket wheel stacker and reclaimer, and plans the single cutting amount with an adaptive layering method, and combines the linkage control with the back-end conveying equipment to effectively improve the stability and efficiency of unmanned stacking operations; at the same time, dynamic monitoring is used to perform real-time analysis of material surface deformation and conveyor belt load information, thereby automatically adjusting the pitch angle, cutting speed and transfer sequence, significantly reducing the risk of bucket wheel overload or material collapse; when foreign matter or torque abnormality is detected, the system can shut down urgently and alarm, realizing safer and continuous bulk material transportation and yard management. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0039] Figure 1 This is an overall flow chart of a method for unmanned automatic stacking and transporting of bulk materials provided by one embodiment of the present invention;

[0040] Figure 2 A model diagram of a bucket wheel reclaiming assembly for an unmanned automatic bulk material stacking and transfer method provided by one embodiment of the present invention;

[0041] Figure 3 A physical diagram of a bucket wheel reclaiming assembly for an unmanned automatic bulk material stacking and transfer method provided by one embodiment of the present invention;

[0042] Figure 4 A physical picture of a bulk material stacking and transfer vehicle for an unmanned automatic stacking and transfer method of bulk materials provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0043] To make the above-mentioned objects, features, and advantages of the present invention more clearly understood, the following detailed description of the specific embodiments of the present invention is given in conjunction with the accompanying drawings. It is obvious that the described embodiments are only part of the embodiments of the present invention, not all of them. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in this field without creative work should fall within the scope of protection of the present invention.

[0044] Example 1, reference Figure 1 - Figure 4, as one embodiment of the present invention, provides an unmanned automatic stacking and transporting method for bulk materials, comprising:

[0045] S1: Collect 3D point cloud data of the bulk material yard, obtain the overall point cloud of the yard based on the 3D point cloud data, and create an elevation map based on the overall point cloud of the yard.

[0046] Furthermore, multiple sensors, such as lidar (360° scanning or multi-beam) and depth cameras, are installed on the bulk material stacker's boom, tower, or bulk material yard roof, where it's easy to overlook the yard. This allows for real-time acquisition of a 3D point cloud of the bulk material yard. Gyroscopes, inclination sensors, and other posture monitoring devices are also deployed in key areas of the vehicle. When the stacking and transfer task begins, the sensors are activated to collect 3D point cloud data from multiple angles. This data includes a point cloud of the material pile surface, ground reference data, and a point cloud of the bulk material stacker's surface.

[0047] Furthermore, the overall point cloud of the storage yard is obtained by stitching the 3D point cloud data. Outliers are removed and downsampled (such as based on VoxelGrid or uniform grid) for each frame of the local point cloud to reduce noise and data volume. Based on the fuselage posture, rotation angle and sensor calibration parameters recorded during sensor acquisition, a preliminary coordinate transformation is first performed on each frame of the point cloud to roughly align it to the same world coordinate system.

[0048] Compute feature descriptors (such as FPFH, ISS feature points, or other geometric features) for the point clouds to provide a feature matching basis for subsequent stitching. Roughly align the point clouds using RANSAC-based feature matching or other matching algorithms to find similar feature pairs and estimate the rotational and translational relationship between the two point clouds in 3D space. Fine-tune the coarse registration results using fine-tuning algorithms such as Iterative Closest Point (ICP) or other local optimization algorithms to minimize point-to-point and point-to-surface errors and ensure more accurate alignment of overlapping areas.

[0049] Repeat the coarse and fine registration processes to align and merge multiple frames of 3D point cloud data into the same world coordinate system to obtain the overall point cloud of the storage yard.

[0050] Furthermore, for more efficient motion planning, the overall point cloud model is typically converted into a discrete 2.5D or 3D representation for algorithmic convenience. In most bulk material storage applications, the material yard exhibits a predominantly undulating terrain (i.e., the horizontal area varies significantly more than the height). Therefore, the top surface of the mesh is extracted to generate an elevation map for pathfinding.

[0051] The specific steps of establishing the elevation map based on the overall point cloud of the storage yard include: setting the resolution of the grid on the plane (x, y) (for example, 1m×1m per grid, or finer or coarser) according to the size of the storage yard and the required accuracy, and calculating the height of the entire storage yard in [x min , x max ] and [y min ,y max ] grid index within the range;

[0052] For each point (x i ,y i , z i ), find the corresponding grid cell (I, J), and count all the z values that fall into the grid point in the cell. i Value, take the maximum height z max (I, J) is the height of the grid cell on the elevation map;

[0053] After completing the calculation of the maximum height of all grids, a two-dimensional matrix corresponding to the grid can be obtained, forming an elevation map.

[0054] Using a 2.5D elevation map (HeightMap) as the main input for motion planning avoids the uncertainty caused by special scenarios such as multi-layer structures or voids. In bulk yards with relatively simple material yard shapes, this surface elevation projection can efficiently support the search and optimization of automatic stacking and transfer paths.

[0055] For typical bulk material handling scenarios without obvious overhead stockpiles (such as coal and ore stockpiles, and cement feedstock), the proposed method enables rapid modeling, real-time updates, and automated planning, significantly improving the efficiency and safety of stockpiling and handling in unmanned environments. For extremely complex 3D structures, a full 3D voxel (VOXEL) or triangular mesh can be used. However, in the bulk material stockpiles targeted by this invention, elevation maps are sufficient for most needs, offering the advantages of greater efficiency and simplified maintenance.

[0056] Furthermore, after obtaining the elevation map, the stockpile is layered in the height direction to guide the bucket wheel reclaimer to cut layer by layer (or segment by segment). In the existing technology, "equal thickness" or "simple multi-level settings" are usually used to segment the stockpile. In fact, the layer thickness, cutting priority, and bucket wheel workload are closely related to the ease of cutting the material. If the material properties can be retrieved from the database before the task begins, the subsequent automatic planning will be more targeted. For example, materials with high looseness can allow a larger single cutting depth, and hard materials require a smaller layer thickness to prevent overload or excessive wear.

[0057] In bulk material storage yards, such as industrial or port yards, common bulk materials such as coal, ore, cement raw materials, and grain often have relatively regular physical properties (hardness, density, moisture content, etc.). Establishing a material attribute database to record the types of bulk materials that are frequently transported and their corresponding hardness, loose density, or moisture content in advance can reduce the workload of temporary on-site measurements. Before starting a stacking and transfer task, the system operation interface allows users to select the "material type for this operation" and automatically call the corresponding attributes in the database, or make parameter corrections before entering the task execution.

[0058] Specifically, a set of key attributes is set for each common bulk material (such as coal A, ore B, grain C, limestone D, etc.): hardness grade (such as Mohs hardness range or compressive strength index), bulk density, safe slope threshold, load on the hopper wheel, etc. The key attribute ranges of each material are obtained based on the historical data of the material, and a material database is established;

[0059] When the staff creates a stacking and transfer task, the operator selects or enters the corresponding material type on the interface. According to the material type entered by the operator, the key attributes of the corresponding material type in the material database are called, and the adaptive layer thickness is calculated to layer the elevation map.

[0060] The layering of the elevation map includes obtaining the overall height range of the yard based on the elevation map and generating a reference layer thickness based on the key attributes of the material. The reference layer thickness is a fixed initial value based on the overall height range of the yard and the material characteristics, through empirical values or verified in previous operations (for example, 2 meters for loose materials and 1 meter for hard materials).

[0061] The slope of each grid cell is read from the elevation map. For each grid cell, the local layer thickness is calculated based on the slope and the hardness of the material, expressed as:

[0062] Δh local (i) = Δh base -α s ·f s (S(i))-α m ·f m (H m )

[0063] Where Δh local (i) represents the local layer thickness of grid cell i; Δh base represents the base layer thickness; α s Indicates the slope correction coefficient; f s (S(i)) represents the slope mapping function; S(i) represents the slope of grid cell i; α m Indicates the hardness correction factor; f m (Hm ) represents the hardness mapping function; H m Indicates the hardness of material m.

[0064] The slope and hardness are mapped to 0 or positive values through the slope mapping function and the hardness mapping function, and the quantification of Δh base The reduction intensity is expressed as:

[0065] f s (S(i))=max(0, S(i)-S safe )

[0066] f m (H m )=max(0,H m -H0)

[0067] Among them, max means taking the maximum value; S safe Indicates the safety slope threshold, which is determined based on the material type; H0 indicates the baseline hardness.

[0068] To prevent the calculated result from the formula from being too high or too low, the calculated Δh local (i) Perform amplitude limiting processing, and set Δh local (i) Limited to the minimum layer thickness Δh that can be effectively cut by the machine min and the layer thickness Δh that ensures that there is no risk of overloading due to excessive thickness in one cutting max The actual layer thickness for each area is obtained.

[0069] It should be noted that the actual layer thickness can be understood as the vertical depth that the bucket wheel needs to dig in one cutting (or one operation cycle) in the area. When the system determines the actual layer thickness Δh of a certain area local (i) means that the bucket wheel excavator will cut off this Δh on the material surface of the current layer in this area. local The height difference is ultimately used to guide the height of the material layer to be cut by the bucket wheel excavator in each operating cycle in the corresponding area, and is the core parameter for realizing the automatic layered cutting strategy of the present invention.

[0070] Slope S(i) or hardness H m When α is large, s ·f s (S(i)) or α m ·f m (H m ) will increase, thereby reducing the local layer thickness Δh local (i) To reduce the excavation volume of the bucket wheel excavator at one time; when the slope and hardness are both small, there are few deduction items in the formula, Δh local (i) Approximate to Δh base , which can quickly cut layers and improve efficiency.

[0071] During the automatic operation process, if the sensor (torque sensor or acoustic vibration sensor) finds that the actual load is too different from the estimated value in the database, the control system can appropriately adjust the layer thickness or lifting degree; if at some point it is identified that the material properties are inconsistent with the pre-registration (such as a sudden increase in density or a sharp increase in moisture content), it can record it in real time and prompt the operator to update the material properties, gradually improving the database.

[0072] Through adaptive layer thickness, the cutting thickness is automatically reduced in high-slope or high-hardness areas to prevent the bucket wheel excavator from exceeding the torque limit or overturning the fuselage. The single layer thickness is maintained or even increased in flat and easy-to-dig areas, speeding up the operation process. Combined with the hardness values of different materials in the database and combined with on-site sensor feedback, the thresholds and coefficients in the formula can be continuously optimized during the task execution, making the system more and more in line with the actual bulk material situation.

[0073] The application of this adaptive stratification concept can make dynamic adjustments based on the local characteristics of the material yard and the bucket wheel excavator's own capabilities, making the material reclaiming process more efficient, stable and safe. Compared with traditional single equal-thickness stratification, although it requires slightly more calculation and sensor information processing, in most large-scale yard automation operations, the efficiency and safety improvements it brings are significantly greater than the additional computing power consumption.

[0074] In operations with commonly used materials, directly calling database parameters can enable the bucket wheel excavator to quickly enter the most suitable cutting and transfer rhythm, reducing blind trial and adjustment time. For high-hardness or highly abrasive materials, pre-tightening the cutting depth or increasing the pitch safety margin can avoid excessive wear of the bucket wheel and the support plate, and also prevent overload or material collapse accidents.

[0075] S2: Collect the body posture data of the bulk material stacking transfer loading vehicle, plan the material reclaiming movement path based on the elevation map according to the body posture data, update the 3D point cloud data in real time and correct the material reclaiming movement path.

[0076] Furthermore, planning the material movement path based on the elevation map includes obtaining the ground height H(i, j) corresponding to each grid cell (i, j) from the elevation map, calculating the slope difference between adjacent grid cells, and marking the grid cell (i, j) as infeasible if the calculated slope difference Slope(i, j) exceeds the safe slope threshold; otherwise, marking it as feasible. Furthermore, physical limitations such as the travel boundaries and turning radius of the bulk material stacker can be taken into account to mark grid cells beyond the reach of the mechanical structure as infeasible.

[0077] The motion planning of the bucket wheel machine is defined on the elevation map as follows: K = {(x, y, θ, φ)}, where (x, y) represents the discrete coordinates of the machine body on the grid, θ represents the orientation angle of the machine body, and φ represents the inclination angle of the bucket wheel.

[0078] Establish constraints: Maximum pitch angle: φ≤φ max ;

[0079] Safe slope: The grids that are too steep have been eliminated in the feasible area determination.

[0080] Fuselage turning radius: On bucket wheel machines with rotating or slewing platforms, it is necessary to ensure that there are no obstacles and collisions in the surrounding area; this can be reflected in the expansion barrier of the feasible grid.

[0081] Height difference of stockpile: reflected in the elevation map. If the elevation of a certain location is too high and exceeds the bucket wheel's cutting capacity, it can also be marked as temporarily unfeasible.

[0082] The sampling space is constructed by the A / D grid method, and the starting position of the bulk material stacking transfer vehicle {(x0, y0, θ0, φ0)} is obtained according to the fuselage posture data. The search is performed, the neighborhood nodes are expanded in sequence, and the cost is accumulated until the material collection position is reached.

[0083] The output material movement path is a node sequence, {(x0, y0, θ0, φ0), (x1, y1, θ1, φ1), ..., (x n ,y n ,θ n ,φ n )}, indicating the trajectory of the fuselage, the change of its orientation and the change of the bucket wheel inclination angle.

[0084] like Figure 4 As shown in the figure, the bulk material stacker transfer vehicle moves on the crawler, and the discrete path (x n ,y n ,θ n ,φ n ), sending walking instructions (how many meters to walk forward / backward) and turning instructions (how many degrees to turn left / right) to the walking drive system in sections to adjust the overall posture of the fuselage.

[0085] like Figure 3 As shown, the bucket wheel assembly is fixed at the front end of the cantilever and can be adjusted up and down, left and right within a certain range through the pitch cylinder and the slewing mechanism. Once the fuselage reference point (x n ,y n ,θ n ,φ n ) is determined, and the bucket wheel position can be calculated based on the geometric relationship. Since the mechanical structure of the fuselage and the bucket wheel remains unchanged, after completing the first calculation of the relative geometric relationship, the bucket wheel position can be directly obtained based on the fuselage position without the need for recalculation.

[0086] It should be noted that in the elevation map, the bulk material stacker occupies a large part of the space. The discrete coordinates (x, y) of the fuselage in the grid are the "motion reference" of the entire machine. The planar motion of the fuselage (including walking and rotation) can be described by its translation and orientation changes. A reference point is selected at the main rotation center of the bulk material stacker (such as the turntable center or the geometric center of the fuselage), and the reference point is defined as the discrete coordinates (x, y) of the fuselage in the grid.

[0087] Furthermore, when performing the bucket wheel reclaiming operation, the bucket wheel inclination angle φ is calculated according to the actual layer thickness to control the bucket wheel reclaiming. The core idea is to use the actual layer thickness to calculate the amount that the pitch cylinder should extend and retract, thereby obtaining the required bucket wheel inclination angle.

[0088] Specifically, the hinge point between the cantilever and the main structure of the fuselage is regarded as the pitch rotation center. The expected drop of the bucket wheel center is determined by taking the difference between the current height of the pitch rotation center and the surface of the material layer and then combining it with the actual layer thickness. According to the inverse kinematics of the single-joint arm in existing mechanics and robotics, the required drop is mapped to the required pitch angle φ n , the control system will φ n The command is sent to the pitch cylinder or motor drive module to drive the arm down (or up) to that angle. In actual machinery, closed-loop control (such as PID or feedback based on the cylinder position sensor / tilt sensor) is often used to continuously correct the output until φ n consistent and stable within the error range.

[0089] Before the bulk material stacker reaches the predetermined reclaiming position, it is necessary to determine the rotation speed or feed speed of the bucket wheel within the thickness range. Specifically, the system first reads the maximum available rotation speed of the bucket wheel and the pitch range allowed by the equipment, and combines the material properties, hardness, and layer thickness of the material yard; sets a target reclaiming rate q (tons / hour) to ensure that the "single layer cutting volume × rotation speed" is neither overloaded nor too conservative, and calculates the corresponding bucket wheel speed. If the forward speed of the fuselage is also required, the volume flow rate can be approximated by the cross-sectional area × forward speed, and then the bucket wheel speed is matched. The target reclaiming rate q can be set to a preset value for the material type and saved as a key attribute in the material database.

[0090] Furthermore, the correction of the material-grabbing movement path includes automatically defining a region of interest (ROI) according to the bucket wheel radius, rotation range and current cutting layer height during bucket wheel operation, such as using the bucket wheel center as the center of the circle with a radius slightly larger than the distance to the outer edge of the bucket wheel, or performing a key scan in the front fan-shaped area, thereby reducing the amount of data processing in irrelevant areas and improving real-time performance. The three-dimensional point cloud data of the area of interest is obtained in real time through the sensor, and the elevation map is locally updated in conjunction with the real-time collected fuselage posture data.

[0091] Retrieve the height distribution H recorded in the previous cycle of the material collection location on the elevation map old (i, j), and the new height distribution H after local update new (i, j) to calculate the difference and obtain the height difference ΔH(i, j). The material removal situation is determined based on the height difference and the actual layer thickness:

[0092] Excessive material extraction: If ΔH(i, j) is greater than the expected material extraction thickness (for example, it exceeds the actual layer thickness by a large margin), it may indicate that material collapse has occurred or that the area has been excavated too deep.

[0093] Insufficient material: If ΔH(i, j) is much smaller than expected, it means that the material surface is still too high and the bucket wheel may not penetrate deep enough.

[0094] Abnormal protrusion: If H is detected new If (i, j) is higher than the actual layer thickness estimate, or foreign matter appears (hardness greater than the set value), it may indicate a solid obstacle or accumulation of collapsed materials.

[0095] The upper and lower limits of deformation are defined based on the layered cutting thickness and the allowable fluctuation range of the material surface.

[0096] If ΔH(i, j) exceeds the upper limit of deformation, it is judged as a "material collapse or over-excavation" situation, and the pitch actuator is commanded to raise the bucket wheel or slow down the speed;

[0097] If ΔH(i, j) is lower than the lower limit of deformation, it is judged that “material is insufficient” and the pitch actuator is commanded to move the bucket wheel down a certain angle or distance;

[0098] If an abnormal protrusion appears, it will be marked as "obstacle avoidance or re-planning required".

[0099] S3: Communicates with back-end transfer equipment to collaboratively control the material retrieving movement path, dynamically calculates the remaining material quantity in the yard based on real-time updated 3D point cloud data, and completes unmanned automatic stacking and transfer of bulk materials.

[0100] Furthermore, the collaborative control includes obtaining the transfer rate and material flow of the rear-end transfer equipment, calculating the material collection rate based on the real-time collected fuselage posture data, and performing collaborative control based on the comparison result of the transfer rate and the material collection rate.

[0101] like Figure 2 As shown, a conveyor belt is connected to the rear of the bucket wheel for transporting materials. The back-end transfer equipment includes a conveyor belt and other subsequent material transfer equipment. Bidirectional data communication is established between the bucket wheel control system and the back-end transfer equipment through industrial Ethernet, wireless network or field bus (such as PROFIBUS / Modbus, etc.). The interface receives the belt speed information and load status of the back-end belt conveyor, and can also send the current material reclaiming speed and estimated material flow of the bucket wheel to the other party.

[0102] Real-time collection of conveyor belt speed v belt : The back-end equipment sensor or inverter feedbacks the current belt speed;

[0103] Belt surface height or material flow l belt : The conveyor belt is equipped with a detection device such as a material level sensor or a belt scale to detect whether the belt surface height exceeds the set threshold;

[0104] Bucket wheel reclaiming rate q new : It can be calculated from the bucket wheel speed and the estimated single cutting amount / layer thickness.

[0105] The coordinated control includes but is not limited to belt speed-reclaiming rate matching, belt height monitoring and abnormal load protection. Specifically, the belt speed-reclaiming rate matching includes: new , after calculation and the current belt speed v belt For comparison:

[0106] If q new >βv belt , indicating that the bucket wheel feeds faster than the belt conveyor's safe transport capacity, a deceleration instruction is sent to the bucket wheel control or the reclaim depth is reduced;

[0107] If q new >γl belt , indicating that the rear end belt speed is still sufficient, and the bucket wheel speed can be appropriately increased to improve efficiency.

[0108] Among them, β and γ are adjustable coefficients, which ensure that the belt load is maintained within a safe range while maximizing the transportation capacity.

[0109] The belt surface height monitoring includes: when the rear end detects the belt surface height l belt If the preset safety threshold is exceeded, the system will automatically send an alarm and control request: the bucket wheel excavator control system will immediately reduce the bucket wheel speed or suspend the excavation action until the belt surface returns to a reasonable range; if there is still no improvement after a period of time, the remote monitoring or central control personnel will be prompted to intervene manually.

[0110] The abnormal load protection includes that if the belt conveyor fails (overload shutdown or deceleration), the information is transmitted to the bucket wheel machine with high priority in the communication interface: after receiving it, the bucket wheel machine immediately stops (or slows down) the material retrieving rotation to ensure that the material does not accumulate at the conveying end and cause blockage or spillage.

[0111] The coordinated control of the bucket wheel excavator and back-end transfer equipment enables intelligent linkage of the entire automatic stacking and transfer system, improving the continuity and stability of bulk material transportation. Through two-way data communication, the bucket wheel excavator can obtain real-time information on the conveyor belt speed, belt height, and load, and match it with its own reclaim rate, dynamically coordinating the reclaim rate with the back-end conveying capacity, avoiding conveyor belt overload caused by excessive feeding or inefficient operation of the conveying system due to insufficient feeding.

[0112] Furthermore, the dynamic calculation of the remaining material volume in the storage yard based on the real-time updated three-dimensional point cloud data includes installing a belt scale or flow sensor inside the machine or on the conveyor belt at the end of the cantilever to measure the instantaneous material conveying volume and calculate the total transfer volume within a period of time Δt by integration:

[0113]

[0114] Among them, P represents the total transport volume; t0 represents the starting time; and p(t) represents the instantaneous transport volume.

[0115] The elevation map can be used to estimate the initial volume and current remaining volume of the layer in the yard. The difference between the two should roughly match P (within the error threshold). If the difference is too large, it indicates that the model does not match the actual situation (perhaps collapsed material has overflowed into other areas or the material density estimate in the database is inaccurate). This can be corrected when the material database or elevation map is updated.

[0116] When it is detected that the remaining material in the layer or area is lower than the threshold (for example, less than 5% of the remaining material), it is considered that the local area is basically empty. The system will give the "next area or lower layer" instruction to let the bucket wheel excavator move to the adjacent layer or the next high point area according to the established layer order until the material removal task of all areas is completed and a task completion signal is sent.

[0117] The system uses material level sensors and belt scales to monitor the conveyor belt's load status, ensuring it remains within a safe range. Once the threshold is exceeded, the system automatically reduces the bucket wheel excavator's reclaim rate or halts operations to prevent spillage or blockage. Furthermore, if the conveyor belt malfunctions due to overload, malfunction, or an emergency stop, this information is quickly transmitted to the bucket wheel excavator's control system via an industrial communication interface, ensuring the excavator immediately adjusts its operating status. This reduces operational interruptions caused by conveyor system failures and improves system stability.

[0118] Example 2, reference Figure 2 -and Figure 3 , which is an embodiment of the present invention, is scientifically demonstrated through simulation experiments in order to verify the beneficial effects of the present invention.

[0119] like Figure 2 and Figure 3The figure shows a model diagram of the bucket wheel reclaiming assembly of the present invention, including a reclaiming bucket wheel, a conveyor belt and related structures.

[0120] The bucket wheel is the core reclaiming component of the equipment. It is equipped with multiple buckets, each with a radial opening and a rotational opening. These buckets are used to grab bulk material during rotation. Claws on the outer edges of the buckets cut and grab the material, forcing it into the buckets and carrying it to the top discharge position as the wheel rotates.

[0121] A conveyor belt is connected to the rear of the bucket wheel to receive the material discharged from the bucket wheel and transport it to the subsequent transfer equipment. The material on the bucket wheel falls evenly onto the conveyor belt through the material guide assembly, ensuring that the material does not scatter or get stuck during the transfer process.

[0122] The bucket wheel's rotation speed, reclaim depth, and pitch angle are dynamically adjusted based on a real-time, updated elevation map. The system first calculates the remaining material in the stockpile and the current layer thickness. Based on the geometric relationship between the bucket wheel and the machine body, it automatically calculates the required pitch angle, ensuring the bucket wheel accurately penetrates the target material layer. During operation, sensors continuously monitor changes in the material surface. If material collapse, undercutting, or overexcavation are detected, the system automatically adjusts the bucket wheel's inclination, rotation speed, or feed rate to ensure operational stability and minimize equipment wear and energy consumption.

[0123] In addition, the bucket wheel excavator is also linked with the back-end conveying system for control, receiving data such as the conveyor belt speed, load status and material surface height through the industrial communication interface, and feeding back its own material collection rate to the conveyor belt system to ensure that the material collection speed matches the conveying capacity. When the conveyor belt surface height exceeds the safety threshold, the system automatically adjusts the bucket wheel speed or suspends operation to prevent the conveyor belt from overloading or spilling material; if the belt machine fails or stops in an emergency, the bucket wheel will also stop collecting material synchronously to ensure the safety and stability of the conveying process. Overall, the bucket wheel collecting assembly of the present invention realizes efficient, stable and unmanned stacking and transportation of bulk materials through intelligent control, precise material collection, dynamic adjustment and conveying linkage, and is suitable for the efficient operation requirements of large-scale automated material yards.

[0124] If the above functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, and other media that can store program code.

[0125] The logic and / or steps represented in the flowcharts or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing the logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (e.g., a computer-based system, a system including a processor, or other system that can fetch and execute instructions from an instruction execution system, apparatus, or device). For purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport a program for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0126] More specific examples (a non-exhaustive list) of computer-readable media include the following: an electrical connection with one or more wires (electronic devices), a portable computer disk cartridge (magnetic devices), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), a fiber optic device, and a portable compact disc read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering, or processing in another suitable manner as necessary, and then stored in a computer memory.

[0127] It should be understood that various parts of the present invention can be implemented using hardware, software, firmware, or a combination thereof. In the above-described embodiments, multiple steps or methods can be implemented using software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented using hardware, as in another embodiment, any one of the following technologies known in the art or a combination thereof can be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application-specific integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

[0128] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for unmanned automatic stacking and transporting of bulk materials, characterized in that: include: Collect 3D point cloud data of the bulk material yard, obtain the overall point cloud of the yard based on the 3D point cloud data, and create an elevation map based on the overall point cloud of the yard; Collect the posture data of the bulk material stacking transfer vehicle, plan the material reclaiming movement path based on the elevation map based on the posture data, update the 3D point cloud data in real time and correct the material reclaiming movement path; It communicates with the back-end transfer equipment, coordinates the material-retrieving movement path, dynamically calculates the remaining material quantity in the yard based on the real-time updated 3D point cloud data, and completes unmanned automatic stacking and transfer of bulk materials.

2. The unmanned automatic stacking and transporting method for bulk materials according to claim 1, characterized in that: The step of obtaining the overall point cloud of the storage yard by stitching the three-dimensional point cloud data includes filtering the three-dimensional point cloud data, performing preliminary coordinate conversion on each frame of the filtered three-dimensional point cloud data, and aligning them to the same world coordinate system. Calculate the feature descriptors of 3D point cloud data, use the RANSAC-based feature matching algorithm for coarse registration, and fine-tune the coarse registration results through fine registration to accurately align the overlapping areas; Repeat the coarse and fine registration processes to align and merge multiple frames of 3D point cloud data into the same world coordinate system to obtain the overall point cloud of the storage yard.

3. The method for unmanned automatic stacking and transporting of bulk materials according to claim 2, characterized in that: The step of establishing an elevation map based on the overall point cloud of the storage yard includes setting a grid resolution on the plane (x, y) based on the overall point cloud of the storage yard, taking (x, y) as plane coordinates, dividing the grid cells according to the resolution, recording the maximum height of each grid cell, and obtaining an elevation map with equal spacing grids; Obtain historical data on bulk material transfers at a bulk material storage yard and assign a set of key attributes to each material, including material hardness grade, bulk density, and safe slope threshold. Obtain the key attribute range of each material based on the historical data of the material and establish a material database; When the staff creates a stacking and transfer task, the corresponding material type and key attributes in the material database are called according to the material type entered by the operator, and the elevation map is layered.

4. The method for unmanned automatic stacking and transporting of bulk materials according to claim 3, characterized in that: The layering of the elevation map includes obtaining the overall height interval of the yard according to the elevation map. min , z max ], generate benchmark layer thickness based on key material attributes; The slope of each grid cell is read from the elevation map. For each grid cell, the local layer thickness is calculated based on the slope and the hardness of the material, expressed as: Δh local (i)=Δh base -α s ·f s (S(i))-α m ·f m (H m ) Where Δh local (i) represents the local layer thickness of grid cell i; Δh base represents the base layer thickness; α s Indicates the slope correction factor; f s (S(i)) represents the slope mapping function; S(i) represents the slope of grid cell i; α m Indicates the hardness correction factor; f m (H m ) represents the hardness mapping function; H m Indicates the hardness of material m; The slope and hardness are mapped to 0 or positive values through the slope mapping function and the hardness mapping function, and the quantification of Δh base The reduction strength of the calculated Δh local (i) Perform clipping to obtain the actual layer thickness for each region.

5. The unmanned automatic stacking and transporting method for bulk materials according to claim 4, characterized in that: Planning the material movement path based on the elevation map includes obtaining the ground height H(i, j) corresponding to each grid cell (i, j) according to the elevation map, calculating the slope difference between adjacent grid cells, and marking the grid cell (i, j) as infeasible if the calculated slope difference Slope(i, j) exceeds a safe slope threshold; The fuselage posture data includes the fuselage heading angle and bucket wheel inclination angle. The bucket wheel machine motion planning is defined on the elevation map as follows: K = {(x, y, θ, φ)}, where (x, y) represents the discrete coordinates of the fuselage on the grid, θ represents the fuselage heading angle, and φ represents the bucket wheel inclination angle. The sampling space is constructed by the grid method. The starting position of the bulk material stacking transfer vehicle {(x0, y0, θ0, φ0)} is obtained according to the fuselage posture data. The search is performed, the neighborhood nodes are expanded in sequence, and the cost is accumulated until the material collection position is reached. The output material picking motion path is a node sequence, {(x0, y0, θ0, φ0), (x1, y1, θ1, φ1), ..., (x n ,y n ,θ n ,φ n )}, indicating the trajectory of the fuselage, the change of its orientation and the change of the bucket wheel inclination angle.

6. The unmanned automatic stacking and transporting method for bulk materials according to claim 5, characterized in that: The correction of the material reclaiming motion path includes, during bucket wheel operation, acquiring three-dimensional point cloud data of the material reclaiming position in real time through sensors, coordinating it with the real-time collected fuselage posture data, and locally updating the elevation map; Retrieve the height distribution H recorded in the previous cycle of the material collection location on the elevation map old (i, j), and the new height distribution H after local update new (i, j) to calculate the difference and obtain the height difference ΔH(i, j). The material removal situation is determined based on the height difference and the actual layer thickness: If ΔH(i, j) is greater than the actual layer thickness, it is judged as excessive material extraction; If ΔH(i, j) is less than the actual layer thickness, it is judged that insufficient material is taken; The actual layer thickness of the next cycle is adjusted according to the material taking situation.

7. The unmanned automatic stacking and transporting method for bulk materials according to claim 6, characterized in that: The collaborative control includes obtaining the transfer rate and material flow of the rear-end transfer equipment, calculating the material collection rate based on the real-time collected fuselage posture data, and performing collaborative control based on the comparison result of the transfer rate and the material collection rate.

8. An unmanned automatic stacking and transporting system for bulk materials, applied to an unmanned automatic stacking and transporting method for bulk materials according to any one of claims 1 to 7, characterized in that: include, The scene module collects 3D point cloud data of the bulk material yard, splices the 3D point cloud data to obtain the overall point cloud of the yard, and creates an elevation map based on the overall point cloud of the yard; The path module collects the posture data of the bulk material stacking transfer vehicle, plans the material reclaiming movement path based on the elevation map based on the body posture data, updates the 3D point cloud data in real time, and corrects the material reclaiming movement path; The collaborative module communicates with the back-end transfer equipment to collaboratively control the material retrieving movement path, dynamically calculates the amount of remaining materials in the yard based on the real-time updated three-dimensional point cloud data, and completes unmanned automatic stacking and transfer of bulk materials.

9. A computer device comprising: memory and processor; The memory stores a computer program, characterized in that when the processor executes the computer program, the steps of the method for unmanned automatic stacking and transporting of bulk materials as described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for unmanned automatic stacking and transporting of bulk materials as claimed in any one of claims 1 to 7 are implemented.

Images