Automatic control method and system for unmanned bucket wheel machine

Abstract

The invention provides an unmanned bucket wheel machine automatic control method and system, and relates to the technical field of unmanned bucket wheel machines, and the method comprises the steps: obtaining a grabbing task demand, a three-dimensional operation area model, the real-time position and attitude data of an unmanned bucket wheel machine, the distribution state information of a material pile, and the operation state data of transportation equipment; path planning processing is conducted according to the grabbing task requirement, the three-dimensional operation area model and the real-time position and attitude data, and a preliminary grabbing path is generated; carrying out transportation collaborative planning according to the operation state data and the preliminary grabbing path to obtain a collaborative grabbing scheme; performing collapse area prediction according to the distribution state information to obtain a prediction result; and the cooperative grabbing scheme is adjusted according to the prediction result, and the collapse area with the risk degree higher than the threshold value is dynamically avoided, so that a grabbing operation scheme is obtained. According to the method, through the dynamic path planning and the real-time updated three-dimensional operation area model, the optimal grabbing path can be quickly generated, and the operation sequence can be adjusted in real time.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned bucket wheel machines, and in particular to an automatic control method and system of an unmanned bucket wheel machine. Background Art

[0002] With the rapid development of industrial automation and intelligent technology, unmanned operation has become an important research direction in the field of large-scale industrial equipment control. In bulk material yards, bucket wheel excavators are the core equipment for material grabbing and transportation. Their control methods traditionally rely mainly on manual operation or semi-automatic technology, but these methods face many problems in practical applications. First, manual operation is inefficient. Operators need to monitor the complex working environment for a long time and manually adjust the grabbing path. Not only is the labor intensity high and the efficiency low, but the grabbing accuracy is also severely limited by manual experience. Secondly, the environmental complexity of the material yard puts forward high requirements for grabbing path planning and equipment stability. In particular, during the grabbing process, the dynamic deformation of the material pile may cause path failure or equipment damage, and traditional technologies are difficult to respond in real time. Although existing automatic control technologies have introduced path planning and sensor feedback mechanisms, they usually rely on fixed path planning and static scene modeling, and can only operate when the working environment changes slightly. Once the dynamic collapse of the material pile, complex path interference or other unpredictable factors are encountered, these systems are difficult to quickly adjust the grabbing plan and path planning, which in turn affects the working efficiency and system safety, and lack effective support for the collaborative ability of transportation equipment.

[0003] Based on the above-mentioned shortcomings of the prior art, there is an urgent need for an unmanned bucket wheel machine automatic control method and system. Summary of the invention

[0004] The purpose of the present invention is to provide an unmanned bucket wheel machine automatic control method and system to improve the above problems. In order to achieve the above purpose, the technical solution adopted by the present invention is as follows:

[0005] In a first aspect, the present application provides an automatic control method for an unmanned bucket wheel machine, comprising:

[0006] Obtain the requirements of the grabbing task, the three-dimensional operation area model, the real-time position and posture data of the unmanned bucket wheel machine, the distribution status information of the material pile, and the operation status data of the transportation equipment;

[0007] Performing grabbing path planning processing according to the grabbing task requirements, the three-dimensional operating area model and the real-time position and posture data, and generating a preliminary grabbing path by analyzing the operating area of ​​the bucket wheel machine, the target position of the grabbing task and the dynamic posture of the bucket wheel machine;

[0008] Performing coordinated transportation planning based on the operating status data and the preliminary grabbing path, dynamically adjusting the operation sequence and path planning of the bucket wheel excavator by analyzing the real-time position, operating speed and load status of the transportation equipment, and obtaining a coordinated grabbing plan;

[0009] The collapse area is predicted according to the distribution state information, and the prediction result is obtained by analyzing the morphology and structure information of the material pile and simulating the influence of different grabbing schemes on the morphology change of the material pile;

[0010] The collaborative grasping scheme is adjusted according to the prediction result, the grasping path is optimized, the collapsed area with a risk level higher than a threshold is dynamically avoided, and a specific grasping action strategy is generated to obtain a grasping operation plan.

[0011] In a second aspect, the present application also provides an unmanned bucket wheel machine automatic control system, comprising:

[0012] The acquisition module is used to obtain the grasping task requirements, the three-dimensional operation area model, the real-time position and posture data of the unmanned bucket wheel machine, the distribution status information of the material pile, and the operation status data of the transportation equipment;

[0013] A planning module, used for performing grabbing path planning processing according to the grabbing task requirements, the three-dimensional operating area model and the real-time position and posture data, and generating a preliminary grabbing path by analyzing the operating area of ​​the bucket wheel machine, the target position of the grabbing task and the dynamic posture of the bucket wheel machine;

[0014] An analysis module is used to perform transportation collaborative planning according to the operation status data and the preliminary grabbing path, and dynamically adjust the operation sequence and path planning of the bucket wheel excavator by analyzing the real-time position, operation speed and load status of the transportation equipment to obtain a collaborative grabbing plan;

[0015] A prediction module, used to predict the collapse area according to the distribution state information, analyze the morphology and structure information of the material pile, and simulate the influence of different grabbing schemes on the morphology change of the material pile to obtain the prediction result;

[0016] The optimization module is used to adjust the collaborative grasping scheme according to the prediction results, dynamically avoid the collapsed area with a risk level higher than a threshold by optimizing the grasping path, and generate a specific grasping action strategy to obtain a grasping operation plan.

[0017] The beneficial effects of the present invention are:

[0018] Through dynamic path planning and real-time updated three-dimensional working area models, the bucket wheel excavator can quickly generate the optimal grasping path and adjust the working order in real time, which significantly improves the grasping efficiency compared with traditional fixed path planning; the discrete element method is used to dynamically simulate the changes in the shape of the material pile, predict the collapse area that may occur during the grasping process, and optimize the grasping path in combination with the dynamic avoidance algorithm, effectively avoid high-risk areas, and ensure operation safety; by analyzing the real-time operating status of the transportation equipment, the grasping rhythm and working path of the bucket wheel excavator are dynamically adjusted to ensure seamless connection between the grasping operation and material transportation, improve the overall operation efficiency and reduce waiting time. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0020] Figure 1 It is a schematic flow chart of an automatic control method of an unmanned bucket wheel machine described in an embodiment of the present invention;

[0021] Figure 2 It is a schematic structural diagram of an automatic control system for an unmanned bucket wheel machine described in an embodiment of the present invention;

[0022] Figure 3 It is a schematic structural diagram of an automatic control device for an unmanned bucket wheel excavator described in an embodiment of the present invention.

[0023] Markings in the figure: 800, an unmanned bucket wheel machine automatic control device; 801, processor; 802, memory; 803, multimedia component; 804, I / O interface; 805, communication component; 901, acquisition module; 902, planning module; 903, analysis module; 904, prediction module; 905, optimization module. DETAILED DESCRIPTION

[0024] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. The components of the embodiments of the present invention generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present invention provided in the drawings is not intended to limit the scope of the claimed invention, but merely represents selected embodiments of the present invention. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0025] It should be noted that similar reference numerals and letters represent similar items in the following drawings, so once an item is defined in one drawing, it does not need to be further defined and explained in the subsequent drawings. At the same time, in the description of the present invention, the terms "first", "second", etc. are only used to distinguish the description and cannot be understood as indicating or implying relative importance.

[0026] Embodiment 1:

[0027] This embodiment provides an automatic control method for an unmanned bucket wheel excavator.

[0028] See also Figure 1 , the figure shows that the method includes steps S100 to S500.

[0029] Step S100, obtaining the grasping task requirements, the three-dimensional operation area model, the real-time position and posture data of the unmanned bucket wheel machine, the distribution status information of the material pile and the operation status data of the transportation equipment;

[0030] In actual applications, the requirements for grabbing tasks are issued by the upper scheduling system, including the type of target materials, the number of grabs, quality requirements, and operation time constraints, to ensure that the operation of the unmanned bucket wheel machine meets the global production scheduling requirements. The construction of the three-dimensional operation area model relies on the point cloud data captured by the laser radar or high-precision camera, and the algorithm is used to model and extract spatial features to generate a dynamic three-dimensional spatial map reflecting the yard environment, so as to achieve a comprehensive perception of the yard structure, terrain features, and obstacle distribution. At the same time, the real-time position and posture data of the unmanned bucket wheel machine are collected and fused through GPS and IMU sensors to obtain its current three-dimensional coordinates and posture information, laying the foundation for dynamic path planning. The distribution status information of the material pile is scanned in real time by radar or visual sensors to generate a three-dimensional material pile model, and the density estimation algorithm is combined to further analyze the distribution characteristics of the material pile, such as the height, volume, and boundary characteristics of the pile. In addition, the operating status data of the transportation equipment is obtained through the wireless communication network, including the current position information, operating speed, load status, etc. of the transportation equipment, to provide support for transportation collaborative planning.

[0031] Step S200, performing grabbing path planning processing according to the grabbing task requirements, the three-dimensional operating area model and the real-time position and posture data, and generating a preliminary grabbing path by analyzing the bucket wheel machine's operating area, the target position of the grabbing task and the dynamic posture of the bucket wheel machine;

[0032] It is understandable that the requirements of the grabbing task in this step clearly define the specific target areas that need to be grabbed in the material pile (including material type, grab quantity and priority order, etc.), and these requirements become constraints for path planning. Then, based on the three-dimensional operation area model, the range of feasible paths in the operation area is determined by extracting spatial features of point cloud data and analyzing terrain. At the same time, combined with the real-time position and attitude data of the bucket wheel machine (such as three-dimensional coordinates and pitch angle, yaw angle, etc.), the current state of the bucket wheel machine is calibrated through multi-sensor fusion technology to provide an accurate reference for the selection of the starting point of the path.

[0033] Step S300: perform transportation collaborative planning based on the operation status data and the preliminary grabbing path, and dynamically adjust the operation sequence and path planning of the bucket wheel excavator by analyzing the real-time position, operation speed and load status of the transportation equipment to obtain a collaborative grabbing plan;

[0034] It should be noted that this step achieves a high degree of coordination between the bucket wheel excavator and the transportation equipment, and significantly improves the overall efficiency of the system by optimizing the operation sequence and path planning; secondly, by dynamically analyzing the load status and operation trajectory of the transportation equipment, task conflicts and waiting time between grasping and transportation are avoided, and resource utilization is improved; finally, the collaborative planning scheme can adapt to the status changes of the transportation equipment in real time, ensuring the stability and efficiency of the unmanned bucket wheel excavator in complex operating environments.

[0035] Step S400, predicting the collapsed area according to the distribution state information, analyzing the morphology and structure information of the material pile, and simulating the effects of different grabbing schemes on the morphology change of the material pile, to obtain a prediction result;

[0036] It can be understood that this step conducts an in-depth analysis of the morphology and structural information of the material pile, and uses simulation technology to evaluate the dynamic impact of different grasping schemes on the morphology changes of the material pile, so as to identify possible collapse areas in advance and provide a reliable basis for subsequent path optimization and action adjustment.

[0037] Step S500: adjust the collaborative grasping plan according to the prediction result, optimize the grasping path, dynamically avoid the collapsed area with a risk level higher than the threshold, and generate a specific grasping action strategy to obtain a grasping operation plan.

[0038] It should be noted that this step is the decision-making link in the automatic control method of the unmanned bucket wheel machine. Its purpose is to dynamically adjust the collaborative grabbing plan based on the prediction results of the collapse area, optimize the grabbing path, avoid high-risk collapse areas, and generate specific grabbing action strategies, and finally form a safe and efficient grabbing operation plan. This step ensures the adaptability and safety of the unmanned bucket wheel machine in a complex yard environment through path adjustment and action optimization.

[0039] Further, step S200 includes step S210 to step S240.

[0040] Step S210: Perform positioning processing according to the grasping task requirements and the three-dimensional operation area model, and screen out the material pile position and boundary that meet the grasping requirements by clustering the point cloud data of the three-dimensional operation area model, and obtain the positioning data of the target grasping area, the spatial position of the positioning data, and the boundary data;

[0041] Specifically, this step first combines the target material type, grasping priority, and work area restrictions in the grasping task requirements, and converts them into constraints for point cloud data screening, such as the height range, density distribution characteristics, and specific spatial range of the material. Subsequently, the point cloud data in the three-dimensional work area model is preprocessed and analyzed using a DBSCAN clustering algorithm. DBSCAN is a density-based clustering method that can effectively identify high-density areas in point clouds and separate boundary points and noise points. The clustering results reflect the distribution characteristics of the material pile, and the screened high-density clustering areas are potential grasping targets.

[0042] After completing the preliminary clustering, the spatial distribution characteristics of the material pile are further analyzed in combination with the grasping task requirements. For example, by calculating the center point coordinates, height gradient, volume and boundary shape of the cluster area, the cluster area is screened and prioritized, and the areas that do not meet the grasping requirements are screened out. Finally, the location and boundary data of the material pile that meet the grasping task are extracted. These data are output in the form of positioning data of the target grasping area, including the center coordinates of the material pile, the spatial boundary contour and the surface height distribution, providing accurate initial input for subsequent path planning.

[0043] Step S220, performing path starting point optimization processing according to the spatial position and real-time position and attitude data, by constructing a cost function with path distance, attitude adjustment energy consumption and obstacle avoidance constraint as weights, and using a gradient descent algorithm to optimize and calculate the path starting point, to obtain optimized path starting point parameters, the path starting point parameters including path starting point coordinates and attitude parameters;

[0044] Specifically, firstly, according to the position of the target area and the current three-dimensional coordinates, pitch angle, yaw angle and other attitude information of the bucket wheel machine, the geometric relationship and attitude adjustment requirements between the bucket wheel machine and the target area are analyzed. The path starting point optimization constructs a cost function with path distance, attitude adjustment energy consumption and obstacle avoidance constraints as the main factors, which is used to balance the relationship between path length and energy consumption while ensuring the safety of the path starting point. By iteratively optimizing the cost function using the gradient descent algorithm, the spatial position and attitude parameters of the starting point are gradually adjusted to minimize the energy consumption of the path from the starting point to the target area and meet the obstacle avoidance conditions. The final path starting point parameters include the three-dimensional coordinates of the path and the optimized attitude parameters, which can not only reduce the transition time and energy consumption of the bucket wheel machine from the current state to the starting point of the operation, but also provide a high-quality starting point for subsequent path planning. The technical effect is reflected in the efficiency and safety of the path starting point, which not only optimizes the path length and attitude adjustment process, but also reduces the energy consumption and collision risk of equipment operation, laying the foundation for path planning and smooth operation.

[0045] Step S230, planning a local grasping path according to the boundary data and the path starting point parameters, generating path nodes by gridding on the boundary of the target grasping area, and combining the A* algorithm with the gradient field constraint to generate the path, so as to obtain a local grasping path node sequence from the path starting point to the target grasping area;

[0046] It can be understood that this step first uses the boundary data of the target grab area, combined with the three-dimensional operation area model, to divide the target area into a regular grid node structure, and generates discrete nodes for path search through gridding to ensure the computational efficiency and regional coverage of path planning. Subsequently, the path starting point parameters are used as the planning starting point, the path starting point is connected to the grid nodes of the target area, and the A* algorithm is used for path search. The A* algorithm is a classic heuristic path planning method that can quickly calculate the optimal path from the starting point to the boundary of the target area by combining the path cumulative cost and the heuristic estimation function, while avoiding redundant calculations. In order to further improve the executability of the path, the planning process also introduces gradient field constraints to constrain the path optimization. The gradient field constraint optimizes the node connection of the path by analyzing the mechanical motion characteristics of the bucket wheel machine (such as steering radius, tilt angle limit, etc.) to ensure that the path is physically feasible and smooth. The local grab path finally generated is a series of node sequences that define the trajectory from the path starting point to the target grab area, and ensure that the path not only meets the dynamic operation capability of the bucket wheel machine, but also avoids possible obstacles or unstable areas.

[0047] Step S240, perform path smoothing processing according to the local grasping path node sequence, generate a continuous path function by interpolation, and optimize the control point position in combination with the dynamic constraint of attitude adjustment to obtain a preliminary grasping path, which includes a continuous path function and bucket wheel machine attitude adjustment instructions.

[0048] It should be noted that the path smoothing firstly processes the path nodes continuously through the interpolation algorithm to generate a continuous path function with high smoothness. The interpolation algorithm can not only smooth the discrete nodes on the transition path, but also avoid sharp corners or unnecessary bends in the path by adjusting the position and curvature of the control points. At the same time, in order to ensure the feasibility of the smooth path in actual mechanical operation, this step combines the dynamic attitude adjustment constraints of the bucket wheel machine (such as the physical restrictions of the pitch angle, yaw angle and bucket wheel rotation angle) to further optimize the position of the path control points. The optimization process corrects the interpolation curve through dynamic constraints to ensure that the generated path meets the smoothness requirements while conforming to the motion characteristics and physical operation range of the bucket wheel machine. The final generated preliminary grasping path includes a continuous path function and corresponding bucket wheel machine attitude adjustment instructions. These instructions provide the bucket wheel machine with specific operating parameters (such as steering angle, lifting height and grasping depth, etc.) to ensure that the bucket wheel machine can operate efficiently and smoothly on the grasping path.

[0049] Further, step S300 includes step S310 to step S340.

[0050] Step S310: Performing a transportation equipment status evaluation process based on the operation status data, calculating the idle capacity and dynamic accessibility of the transportation equipment by analyzing the real-time position, operation speed and current load status of the transportation equipment, and obtaining a coordination index of the transportation equipment;

[0051] In this embodiment, the state evaluation of the transport equipment first starts with the real-time position and running speed data, and estimates the moving path of the transport equipment in the short future time and the time window for reaching a specific location through a trajectory prediction algorithm (such as a prediction model based on Kalman filtering). Combined with the grabbing position and path planning results of the bucket wheel machine, the spatial distance and time matching between the transport equipment and the bucket wheel machine are evaluated. At the same time, combined with the current load status of the transport equipment (such as the remaining carrying capacity or the volume of the material bin), the idle capacity of the transport equipment is calculated to determine whether it can undertake the material quantity of the current grabbing task. For the dynamic accessibility of the transport equipment, by combining the speed, acceleration and environmental restrictions of the equipment, a comprehensive evaluation is made as to whether the transport equipment can smoothly reach the target area within the specified time. Finally, by combining the above analysis results, a coordination index of the transport equipment is generated. This index is a quantitative result, including the arrival time prediction, spatial adaptability and load matching of the transport equipment, which serves as an important input basis for transport coordination planning.

[0052] Step S320: According to the coordination index and the preliminary grasping path, the operation sequence is adjusted, and the operation sequence of the grasping tasks is dynamically adjusted by preferentially matching the idle state of the transport equipment and the sequence constraint of the grasping path, in combination with the optimal scheduling algorithm of the transport equipment, to obtain the optimized grasping sequence;

[0053] Specifically, this step first uses the coordination index of the transport equipment (including idle capacity, dynamic accessibility and arrival time window) to prioritize the current state of the transport equipment. For example, when the idle capacity of the transport equipment is sufficient and the time it arrives at the target area is highly matched with the bucket wheel machine's grabbing operation time window, the transport equipment will be given priority. At the same time, according to the order constraint of the preliminary grabbing path, the material pile position and the expected grabbing time corresponding to each grabbing path are analyzed and matched with the dynamic state of the transport equipment. In this way, the optimal correspondence between different transport equipment and grabbing tasks can be determined. In order to achieve optimal scheduling, this step combines the scheduling algorithm of the transport equipment, preferably such as a task allocation algorithm based on heuristic rules or a dynamic programming algorithm, to adjust the operation order of the grabbing task. In the scheduling algorithm, with the overall efficiency of the system as the goal, the grabbing tasks are dynamically reordered by balancing the load capacity of the transport equipment, the path planning of the bucket wheel machine and the transportation time constraint. For example, the equipment with the closest load to the full load but still able to undertake the current grabbing task is prioritized, or the execution order of the grabbing task is adjusted according to the shortest driving path of the transport equipment to ensure seamless connection between transportation and grabbing.

[0054] Step S330: perform path synchronization adjustment processing according to the optimized grabbing sequence, combine the position of the transport equipment with the dynamic time window matching of the bucket wheel excavator grabbing path, adopt the path adjustment algorithm based on time window optimization, generate a synchronous grabbing path matching the operating state of the transport equipment, and obtain a preliminary collaborative path plan;

[0055] Specifically, first, based on the optimized grabbing sequence, the current position of the transport equipment, the estimated time window to reach the target grabbing area, and the dynamic operating status of the transport equipment (such as speed and acceleration) are analyzed. Combined with the current grabbing path of the bucket wheel excavator, the coordination timing between the transport equipment and the bucket wheel excavator in the grabbing area is evaluated through the dynamic time window matching algorithm. The core of time window matching is to predict whether the arrival time of the transport equipment coincides with the time when the bucket wheel excavator completes the grabbing task, so as to avoid the situation where the transport equipment is waiting or the bucket wheel excavator is delayed, resulting in material accumulation.

[0056] In the path synchronization adjustment, the path adjustment algorithm based on time window optimization is used to optimize the grabbing path of the bucket wheel machine. The algorithm dynamically adjusts the node sequence in the preliminary grabbing path to accurately match the timing of the grabbing task with the arrival time of the transportation equipment. For example, when the transportation equipment is expected to arrive late, the execution order of the grabbing path of the bucket wheel machine is adjusted to execute the grabbing tasks with lower priority in advance to fill the idle time; when the transportation equipment arrives early, the path is adjusted to quickly complete the grabbing tasks in the target area. In addition, the motion characteristics of the bucket wheel machine (such as path smoothness and mechanical constraints) are considered during the path adjustment process to ensure that the adjusted path is feasible in actual operation.

[0057] The resulting synchronous grabbing path is a path plan that is dynamically adjusted according to the operating status of the transport equipment, which not only meets the arrival time requirements of the transport equipment, but also ensures the smooth connection of the grabbing tasks.

[0058] Step S340: Perform transport collaborative optimization processing according to the preliminary collaborative path plan, dynamically adjust the operating speed and loading rhythm of the bucket wheel excavator's grabbing path, iteratively optimize the collaborative plan in combination with the load state constraints, and obtain a collaborative grabbing plan.

[0059] In this embodiment, firstly, based on the preliminary collaborative path plan, the real-time operating status of the transport equipment is analyzed, including information such as the current position of the transport equipment, the estimated arrival time, the current load level, and the remaining carrying capacity. Then, the grabbing path execution rhythm of the bucket wheel machine is dynamically adjusted according to the load state constraints of the transport equipment (such as the maximum load capacity and loading speed of the transport equipment). Specifically, when the transport equipment is close to full load, the rhythm of grabbing materials is reduced by controlling the operating speed of the bucket wheel machine to avoid overloading of the transport equipment due to excessive accumulation of materials; and when the idle capacity of the transport equipment is large, the grabbing rhythm and loading rate of the bucket wheel machine are increased to make full use of the carrying capacity of the transport equipment and improve the operating efficiency.

[0060] In order to achieve collaborative optimization, this step uses an iterative optimization algorithm, and in each iteration, the collaborative solution is updated according to the speed adjustment of the bucket wheel grabbing path and the load status of the transport equipment. The optimization goal is to minimize the waiting time between the bucket wheel and the transport equipment, while maximizing the load utilization of the transport equipment and the grab efficiency of the bucket wheel. After each iteration, the operating speed and loading rhythm are updated according to the latest operating status of the transport equipment until the collaborative solution converges to the optimal solution.

[0061] Further, step S400 includes step S410 to step S440.

[0062] Step S410: extracting the morphological features of the material pile according to the distribution state information, extracting the surface gradient, local curvature and boundary morphological features by using normal vector calculation and surface fitting algorithm on the three-dimensional point cloud data of the material pile, and marking the key areas to obtain the three-dimensional morphological feature parameters of the material pile;

[0063] First, a high-precision three-dimensional model of the material pile is constructed through the preprocessing of the three-dimensional point cloud data. Subsequently, the normal vector calculation technology is used to calculate the normal vector distribution of the surface based on the local neighborhood of each point in the point cloud, which is used to characterize the inclination direction and gradient of the surface of the material pile. The surface gradient is one of the important characteristics of the material pile morphology and can be used to identify steep areas or potential slip risk points. Next, the local curvature of the material pile surface is calculated through a surface fitting algorithm (such as polynomial fitting or quadratic surface fitting based on the least squares method). The local curvature reflects the degree of convexity and concavity of the material pile surface, and can effectively identify high convex areas (such as the top of the pile) and low concave areas (such as the bottom accumulation points) in the pile body. These areas are often the key points of changes in the morphology of the material pile.

[0064] In addition, to further identify the boundary morphological features of the material pile, this step combines the spatial distribution information of the point cloud data, extracts the boundary contour of the material pile through the boundary detection algorithm (such as the Alpha shape algorithm), and marks the key areas in the pile where morphological changes may occur, such as high and steep slopes, weak boundaries, etc. The marked key areas are not only used for subsequent collapse risk prediction, but also provide important references for dynamic simulation and grasping path planning.

[0065] The 3D morphological characteristic parameters of the material pile finally extracted include surface gradient, local curvature, boundary morphology and marking information of key areas. These parameters fully describe the geometric characteristics and structural features of the material pile.

[0066] Step S420: Perform structural stability analysis based on the three-dimensional morphological characteristic parameters, construct a stability discrimination model based on morphological characteristics, combine gravity field analysis and contact surface distribution calculation, perform overall stability zoning evaluation on the material pile, and obtain a stability distribution map of the material pile;

[0067] First, according to the three-dimensional morphological characteristic parameters extracted in step S410, a stability discrimination model based on the morphological characteristics of the material pile is constructed. This model is mainly used to identify weak areas and potential unstable areas in the pile, and evaluate the overall and local stability of the pile by combining morphological characteristics with mechanical properties. Subsequently, the stress state inside the material pile is calculated through gravity field analysis. Based on the three-dimensional model of the material pile and the material density parameters, the pile is divided into finite volume units, and the gravity vector of each unit and its distribution on the contact surface are calculated. Gravity field analysis can reflect the stress and force concentration areas of each part of the pile, and combined with the morphological characteristic discrimination model, the stability assessment can be further refined. Next, the contact conditions between the material pile and the ground and the contact characteristics between the particles are analyzed through contact surface distribution calculation. The contact surface distribution calculation adopts the normal force and friction force model, combined with the Coulomb friction condition to evaluate the anti-slip ability between the particles and the support force distribution in the contact area between the pile and the ground.

[0068] Finally, the results of the morphological feature discrimination model, gravity field analysis, and contact surface distribution calculation were combined to calibrate the overall stability of the material pile through a zoning evaluation method. The stability distribution map visualizes the stability levels of different areas of the material pile, where high-stability areas are used for path planning and job priority sorting, while low-stability areas are used as the focus of collapse prediction and dynamic avoidance.

[0069] Step S430, performing morphological change simulation processing according to the stability distribution diagram of the material pile, dynamically simulating different grasping schemes by using the discrete element method, simulating the influence of the grasping path on the material pile structure, predicting the morphological changes caused by the grasping process, and obtaining the dynamic corresponding relationship between the grasping path and the morphological changes of the material pile;

[0070] It is understandable that this step first determines the morphological characteristics and stability level of the key areas based on the stability distribution map of the material pile, and identifies high-risk areas that may be affected by the grab (such as low stability areas or boundary areas with weak support). Combined with the grab path and motion parameters of the bucket wheel, the material pile model is discretized into a simulation system composed of particle units. Each particle unit is assigned physical properties and an initial position, and dynamic calculations are performed using the discrete element method. In the dynamic simulation, the discrete element method gradually iterates and calculates the morphological changes of the material pile by simulating the contact forces (including normal forces, tangential forces, friction forces, and adhesion forces) between particles and the interaction between particles and the bucket wheel grab tool. During the simulation, as the grab path is executed, the motion state of the particles (such as displacement, rotation) and the pile morphology (such as slip, sedimentation, or local collapse) are updated in real time. For example, when the grab path passes through a low stability area, the simulation may predict the expansion of the slip boundary or a significant change in the center of gravity of the pile, and these dynamic changes will directly affect the overall stability of the material pile. By comparing and analyzing the simulation results of different grabbing schemes, a dynamic correspondence between the grabbing path and the change in the shape of the material pile is generated. This relationship is expressed in the form of a data graph or a three-dimensional model, describing the degree and range of the change in the pile shape that each grabbing path may cause.

[0071] Step S440: perform collapse risk prediction processing according to the dynamic correspondence, calculate the risk probability distribution of each area during the grasping action, mark the collapse high risk area on the grasping path to obtain the collapse area prediction result.

[0072] Specifically, first, based on the dynamic correspondence between the grasping path generated in step S430 and the morphological changes of the material pile, the direct impact of the grasping action on the material pile is analyzed. On this basis, the grasping path is divided into several key points or areas, and the risk probability is calculated for each area. The risk probability distribution is comprehensively evaluated by combining the simulation results of morphological changes, the stability distribution diagram of the material pile, and the stress state of the particles. Specifically, weak areas (such as low stability areas or areas with insufficient support) are given higher basic risk weights because their structures are easily destroyed; and the risk probability is further increased at path points where large particle displacements or slip extensions occur in dynamic simulations.

[0073] In the risk prediction model, a method combining probability calculation and mechanical analysis is introduced. Furthermore, based on Coulomb friction theory and particle contact mechanics, the probability of slippage or collapse in each area under the grasping action is calculated. In addition, based on the time series data of dynamic correspondence, the gradual accumulation and evolution trend of risks in the grasping process are predicted, and the key nodes that may trigger chain reactions are identified. High-risk areas are classified and marked according to their predicted probability and the risk threshold set by the system. These marking results are displayed in a visual way to form a collapse area risk map of the grasping path.

[0074] Further, step S420 includes step S421 to step S424.

[0075] Step S421, performing contact surface extraction processing according to the three-dimensional morphological feature parameters, analyzing the material pile point cloud data and normal vector distribution, using the Alpha shape algorithm to triangulate the material pile surface to extract the contact area with contact force, and calculating the normal angle distribution and normal area weight of each contact area to obtain the contact surface distribution characteristics of the material pile;

[0076] It can be understood that the surface of the material pile is first geometrically reconstructed based on the point cloud data and normal vector information of the material pile. After preprocessing, the point cloud data generates a normal vector field of the material pile surface. The normal vector distribution reflects the local directional characteristics and inclination of the material pile surface, providing an important basis for the subsequent identification of the contact surface. Then, the Alpha shape algorithm is used to reconstruct the surface and extract the contact surface of the point cloud data of the material pile. The Alpha shape algorithm is a geometric reconstruction method suitable for processing three-dimensional point cloud data. By adjusting the value of the parameter α, the boundary shape and detail features of the material pile can be accurately captured. By converting the point cloud into a triangulated mesh, the Alpha shape algorithm can effectively identify the geometric relationship between the surface of the material pile and the contact area, and extract the key areas where contact forces exist (such as the contact area between the material pile and the ground, and the local contact surface between particles). After extracting the contact surface, the mechanical properties of the contact surface are further quantified by calculating the normal angle distribution and normal area weight of the contact area. The normal angle distribution reflects the inclination of the contact surface and the direction of force. The area with a larger inclination angle often corresponds to a higher risk of slip. The normal area weight indicates the support capacity of the contact surface. The larger the area and the more evenly distributed the area, the higher the stability. These contact surface distribution characteristics provide high-precision input parameters for subsequent mechanical analysis.

[0077] Step S422, performing gravity distribution calculation processing according to the contact distribution characteristics, by dividing the three-dimensional shape of the material pile into finite volume units, using the finite element method to calculate the center of gravity coordinates and gravity distribution of each unit, and combining the normal characteristics of the contact surface to perform mechanical decomposition of the support force and sliding force of each unit, to obtain the gravity field distribution inside the material pile;

[0078] It should be noted that this step can simplify the calculation process and improve the calculation accuracy by decomposing complex geometric shapes into simple units. The finite element method is used to calculate the center of gravity coordinates and gravity distribution of each unit, which can handle complex boundary conditions and material properties, and determine the stress and displacement state of each unit by solving the equilibrium equation. Combined with the normal characteristics of the contact surface, the support force and sliding force of each unit are mechanically decomposed. This process takes into account the mechanical behavior of the material pile when it contacts the ground or other objects. The normal characteristics reflect the inclination degree and force direction of the contact surface, which is crucial for evaluating the stability of the material pile. Through this decomposition, the gravity field distribution inside the material pile can be calculated more accurately, thereby providing reliable data support for subsequent stability analysis.

[0079] Step S423, performing stability analysis according to the gravity field distribution, calculating the anti-slip force of each area on the contact surface based on the preset Coulomb friction model, calculating the stability factor in combination with the gravity field distribution and the support force of the local contact point, and classifying the stability of each contact surface area according to the critical value to obtain the local distribution of the stability factor;

[0080] It should be noted that the Coulomb friction model is a classic model that describes the friction characteristics of granular materials. It takes into account the internal friction angle and cohesion of the material and can effectively predict the slip behavior between particles. By calculating the anti-slip force on the contact surface, the ability of the material pile to resist slip in different areas can be determined. Then, the stability factor is calculated by combining the gravity field distribution and the support force of the local contact point. The stability factor is an important indicator for measuring the stability of the material pile. It comprehensively considers the relationship between the sliding force under gravity and the anti-slip force on the contact surface. By calculating the stability factor of each contact surface area, the stability of the material pile in different areas can be quantitatively evaluated. Finally, the stability of each contact surface area is classified according to the critical value to obtain the local distribution of the stability factor. The critical value is an important reference value for judging the stability of the material pile. When the stability factor is lower than the critical value, the material pile may be in an unstable state and there is a risk of collapse. By classifying the stability of the contact surface area, the weak links and potential risk areas in the material pile can be identified, providing an important basis for subsequent collapse risk prediction and path planning.

[0081] Step S424: perform zoning processing according to the local distribution of the stability factor, and construct a global evaluation model based on the stability factor. Perform weighted interpolation on the locally distributed stability factor and combine it with the transition continuity between regions. Then, mark the high-risk areas and classify the risk levels to obtain the overall stability distribution map of the material pile.

[0082] Specifically, this step first performs zoning processing according to the local distribution of the stability factor, divides the material pile into different areas, and each area is classified according to the value of its stability factor. Zoning processing helps to focus on those areas with poor stability and high risk. Next, a global assessment model based on the stability factor is constructed, which extends the local stability factor data to the entire material pile through the weighted interpolation method. Weighted interpolation takes into account the relative importance of stability factors in different regions, ensuring the accuracy and reliability of the assessment results. At the same time, combined with the transition continuity between regions, it ensures a smooth transition of the stability assessment in space, avoids sudden changes in the assessment results caused by zoning, and makes the overall stability distribution map more realistically reflect the actual situation of the material pile. Finally, the high-risk areas are marked and risk level classified. This step clearly identifies the areas with the lowest stability and the highest risk in the material pile in a visual way, and provides clear risk warnings and decision-making basis by grading them according to the degree of risk.

[0083] Further, step S430 includes step S431 to step S434.

[0084] Step S431, constructing a grasping action mechanics model according to the stability distribution diagram of the material pile, combining the stability distribution diagram of the material pile with the geometric model of the grasping tool, using a particle contact mechanics algorithm based on the discrete element method, decomposing the material pile into a number of discrete particle units, combining the calculation of contact points, contact forces and friction forces between particles, constructing a grasping action mechanics model of the material pile, and obtaining initial state parameters describing the internal stress of the material pile and the movement of particles;

[0085] It can be understood that the stability distribution map of the material pile provides stability information of each area of ​​the material pile, which provides important basic data for the construction of the model. These stability data are combined with the geometric model of the gripper to ensure that the model can accurately reflect the interaction between the gripper and the material pile. The discrete element method is a numerical simulation technique used to simulate the mechanical behavior of granular materials. By decomposing the material pile into a number of discrete particle units, the interaction between particles can be captured in detail, including contact points, contact forces, and friction forces. The discrete element method can accurately calculate the forces and motion states between particles, thereby constructing a comprehensive mechanical model of the gripping action. This model can describe the stress distribution inside the material pile and the initial state parameters of the particle motion, providing the necessary initial conditions for subsequent dynamic simulations. The specific calculation formula is as follows:

[0086] The total force formula is:

[0087] F i =∑ j∈C(i) (F n,ij +F t,ij +F d,ij )+F g,i +F e,i +F c,i ;

[0088] The formula for normal contact force is:

[0089] F n,ij =k n δ n,ij n ij -η n v n,ij ;

[0090] The formula for tangential contact force is:

[0091] F t,ij =min(k t δ t,ij ,μ|F n,ij |)t ij -η t v t,ij ;

[0092] The collision force formula is:

[0093]

[0094] Rotational dynamics formula:

[0095]

[0096] Where i represents the number of the particle; j represents the number of the particle that particle i contacts; C(i) represents the set of particles in the contact area of ​​particle i; F irepresents the total force on particle i; F n,ij represents the normal contact force between particles i and j; k n represents the normal elastic coefficient between particles; δ n,ij represents the normal overlap between particles i and j; n ij represents the normal unit vector from particle i to particle j; η n represents the normal damping coefficient; v n,ij represents the normal relative velocity between particles i and j; F t,ij represents the tangential contact force between particles i and j; k t represents the tangential elastic coefficient between particles; δ t,ij represents the relative sliding displacement between particles i and j; μ represents the sliding friction resistance between particles; t ij represents the tangential unit vector between particles i and j; η t represents the tangential damping coefficient; v t,ij represents the tangential relative velocity between particles i and j; F c,i represents the collision force of particle i; ξ represents the collision energy dissipation coefficient; U represents the potential energy function; r i represents the position vector of the center of mass of particle i; T i represents the total torque of particle i; r ij represents the displacement vector from the center of mass of particle i to the contact point; η r represents the rotation damping coefficient; ω i represents the angular velocity of particle i; F g,i represents the gravity on particle i; F e,i Represents the external force applied by the gripper; F d,ij represents the viscous damping force between particles i and j.

[0097] Step S432, performing dynamic simulation processing of particle motion according to the initial state parameters, analyzing the particle force at each path point of the grasping path, gradually and iteratively calculating the force balance state and motion trajectory of the particles under the action of the grasping tool, dynamically updating the contact relationship and stacking morphology between the particles, and obtaining the dynamic morphological change of the particle distribution of the material pile due to the grasping path;

[0098] It should be noted that this step performs a detailed particle force analysis for each path point of the grasping path. Specifically, it involves calculating the various forces that the particles are subjected to under the action of the grasping tool, such as gravity, friction, contact force, etc., and how these forces affect the motion state of the particles. Through step-by-step iterative calculations, the simulation model can accurately simulate the force balance state and motion trajectory of the particles during the grasping process. As the grasping tool moves, the contact relationship between the particles will change, such as the formation of new contact points and the breaking of old contact points. The simulation model will update these contact relationships in real time to ensure that the interaction between the particles always accurately reflects the actual situation. At the same time, the stacking morphology of the particles will also change, such as the rearrangement, slippage or local collapse of the particles. By simulating the entire grasping path, the dynamic morphological changes of the particle distribution of the material pile can be obtained, that is, the changes in the position, stacking morphology and interaction relationship of the particles at different time points.

[0099] Step S433, extracting the affected area of ​​the morphology change according to the dynamic morphology change, and performing spatial clustering analysis on the change areas of particle displacement and stacking morphology in the simulation results, identifying the high displacement area, sliding boundary area and local collapse area within the action range of the grasping path, and obtaining the affected area of ​​the morphology change;

[0100] It can be understood that this step first analyzes the displacement and stacking morphology changes of the particles in the simulation results. By monitoring the position changes of the particles and the evolution of the stacking morphology, it is possible to determine which areas have undergone significant displacement or morphological changes during the grasping process. Next, the spatial clustering analysis technique is used to classify these change areas. Spatial clustering analysis can identify areas with similar characteristics in space, thereby extracting areas with large particle displacements, boundary areas where slip occurs, and areas where local collapse occurs. High displacement areas refer to areas where particles move a large distance during the grasping process, and these areas may become the core areas of material pile morphology changes; slip boundary areas are boundaries where relative sliding occurs between particles, and the stability of these areas has an important impact on the stability of the entire material pile; local collapse areas refer to areas where local collapse occurs under the action of grasping, and these areas are weak links in the morphological changes of the material pile.

[0101] Step S434, establish a dynamic correspondence based on the impact area of ​​the morphological change, match and analyze the dynamic characteristic parameters of the impact area with the time series of the grasping path, and use an association modeling method based on time series interpolation and regional influence intensity weighting to generate a grasping dynamic correspondence diagram to obtain the dynamic correspondence between the grasping path and the morphological change of the material pile.

[0102] It can be understood that this step extracts the dynamic characteristic parameters of the affected area of ​​morphological changes, such as particle displacement velocity, slip rate, collapse degree, etc. These parameters can quantitatively describe the changes in the area during the grasping process. Then, these dynamic characteristic parameters are matched and analyzed with the time series of the grasping path. The time series of the grasping path records the position and action state of the grasping tool at different time points. By corresponding the changes in the affected area with the time points of the grasping path, it can be determined which areas have changed significantly at which stage of the grasping process. The dynamic characteristic parameters of the affected area are interpolated by a method based on time series interpolation to fill the data gaps in the time series, making the change process more continuous and complete. At the same time, the regional influence intensity is weighted. Considering the different degrees of influence of different areas on the overall stability of the material pile, the weighting can highlight the impact of changes in key areas on the whole. Through the association modeling method, the above analysis results are integrated to generate a grasping dynamic correspondence diagram. The relationship diagram intuitively shows the dynamic connection between the grasping path and the morphological changes of the material pile, and clarifies the response and change trend of each area of ​​the material pile at each stage of the grasping process. The dynamic correspondence between the grasping path and the changes in the material pile shape obtained provides an important decision-making basis for optimizing the grasping path, avoiding risk areas, and improving work efficiency.

[0103] Further, step S500 includes step S510 to step S540.

[0104] Step S510: Dynamically identify and process high-risk areas according to the prediction results and the collaborative grasping scheme, calculate the spatial overlap of the marked high-risk areas for collapse, use the KD tree to locate the intersection nodes of the grasping path and the collapsed area, and select the areas to be avoided according to the risk level threshold to obtain the conflict node set between the grasping path and the high-risk area;

[0105] It should be noted that the prediction results provide a stability assessment of the material pile under different grasping schemes, while the collaborative grasping scheme clarifies the grasping path and the order of operations. Combining the two, it is possible to determine which areas have a higher risk of collapse under a specific grasping path. Next, a spatial overlap calculation is performed to spatially compare the marked high-risk areas of collapse with the grasping path to find the overlapping parts of the two. This process requires precise spatial positioning and computing power to ensure the accuracy of the recognition results. The KD tree algorithm is used to locate the intersection nodes of the grasping path and the collapse area. The KD tree is an efficient spatial search data structure that can quickly find the nearest neighbor points in space. By constructing a KD tree, the specific nodes where the grasping path intersects with the collapse area can be quickly located, and these nodes are potential risk points. Finally, the identified intersection nodes are screened according to the preset risk level threshold. The risk level threshold is set based on factors such as the stability of the material pile, the possibility of collapse, and the degree of impact on the safety of the operation. Only when the risk level of the intersection node is higher than the threshold, it is included in the area to be avoided. Through this screening process, a set of conflicting nodes between the grasping path and the high-risk area is obtained, which provides a clear goal for subsequent path adjustment and avoidance strategy formulation.

[0106] Step S520, performing path avoidance processing according to the conflict node set, by regenerating an alternative path from the grabbing starting point to the target area in the three-dimensional working area model, defining constraint conditions as avoiding conflict nodes, shortest path priority, and bucket wheel machine posture adjustment restrictions, and obtaining an optimized grabbing path;

[0107] Specifically, in the three-dimensional operation area model, we first need to consider how to generate a new alternative path from the grabbing starting point to the target area. This path must meet several key constraints: First, the path must avoid all identified conflicting nodes to ensure that the bucket wheel machine does not enter the high-risk area during the grabbing process, thereby avoiding potential collapse accidents. Second, follow the principle of shortest path priority to improve work efficiency and reduce energy consumption. On the premise of meeting safety and efficiency, the attitude adjustment restrictions of the bucket wheel machine must also be considered to ensure that the path is executable within the physical and operational capabilities of the bucket wheel machine. This involves the reasonable planning of attitude parameters such as the steering angle and lifting height of the bucket wheel machine to ensure the smoothness and accuracy of the grabbing action. By comprehensively considering these constraints, the path planning algorithm is used to search and generate an optimal alternative path in three-dimensional space. This path not only avoids risks, but also takes into account the efficiency of the operation and the operating characteristics of the bucket wheel machine.

[0108] Step S530, performing grabbing action strategy optimization processing according to the optimized grabbing path, sampling each path point, combining the real-time position and posture data of the bucket wheel excavator, and using the action parameter optimization algorithm based on gradient descent to calculate the grabbing depth, angle adjustment and force control of the grabbing action, and obtain the action strategy corresponding to each position on the grabbing path;

[0109] In this embodiment, the optimized grasping path is first uniformly sampled, and the path is divided into several key points, each of which represents the position that the bucket wheel machine needs to pass during the grasping process. Then, the real-time position and posture data of the bucket wheel machine at each sampling point, including the three-dimensional coordinates, pitch angle, yaw angle, etc. of the bucket wheel machine, provide an accurate reference basis for the optimization of the action parameters. Next, the action parameter optimization algorithm based on gradient descent is adopted. By calculating the gradient of the objective function (such as grasping efficiency, energy consumption, etc.), the action parameters are adjusted along the direction of gradient descent to gradually approach the optimal solution. In this step, the objective function is to maximize the grasping efficiency or minimize the energy consumption. Through the gradient descent algorithm, the action parameters such as grasping depth, angle adjustment and force control are optimized. The optimization of grasping depth can ensure that the bucket wheel machine can accurately grasp the material at different positions; the optimization of angle adjustment helps the bucket wheel machine to maintain stable grasping in complex terrain; the optimization of force control can avoid excessive damage to the material pile or loose grasping. Finally, an optimal action strategy is calculated for each position on the grasping path, including the grasping depth, angle adjustment range, and force that should be performed at that position. These strategies will guide the bucket wheel machine on how to accurately adjust its actions during the actual grasping process to achieve efficient and safe grasping operations.

[0110] Step S540, generate an operation plan based on the grasping action strategy and the optimized grasping path. By integrating the dynamic avoidance path with the grasping action strategy, the path is continuously optimized using cubic B-spline curve interpolation. At the same time, the rhythm of the grasping action is synchronously adjusted in combination with the operating status of the transportation equipment to obtain an optimized grasping operation plan.

[0111] It can be understood that this step first integrates the dynamic avoidance path with the grabbing action strategy to ensure that the grabbing action can be performed smoothly while avoiding risks. The dynamic avoidance path provides a safe path selection, while the grabbing action strategy defines specific grabbing parameters for each path point. The combination of the two makes the grabbing operation safe and efficient. Next, the path is continuously optimized using cubic B-spline curve interpolation. The cubic B-spline curve is a smooth curve interpolation method that can generate a smooth and continuous path, reduce the sharp turns or mutations that may occur during the execution of the bucket wheel machine, and thus improve the stability and smoothness of the grabbing action. At the same time, the rhythm of the grabbing action is synchronously adjusted in combination with the operating status of the transportation equipment. Considering the real-time position, speed and load of the transportation equipment, the rhythm of the grabbing action is reasonably arranged to achieve seamless connection between grabbing and transportation, reduce waiting time and improve overall operation efficiency. Through these comprehensive adjustments, an optimized grabbing operation plan is finally obtained, which not only takes into account safety and efficiency, but also takes into account the collaborative operation with the transportation equipment, providing a strong guarantee for the efficient and safe grabbing operation of the unmanned bucket wheel machine. This step achieves a perfect combination of grasping path, action strategy and transportation coordination by generating a fully optimized grasping operation plan, significantly improving the overall performance of the operation.

[0112] Embodiment 2:

[0113] like Figure 2 As shown, this embodiment provides an unmanned bucket wheel machine automatic control system, the system includes:

[0114] The acquisition module 901 is used to obtain the grasping task requirements, the three-dimensional operation area model, the real-time position and posture data of the unmanned bucket wheel machine, the distribution status information of the material pile, and the operation status data of the transportation equipment;

[0115] Planning module 902, used to plan and process the grabbing path according to the grabbing task requirements, the three-dimensional working area model and the real-time position and posture data, and generate a preliminary grabbing path by analyzing the working area of ​​the bucket wheel machine, the target position of the grabbing task and the dynamic posture of the bucket wheel machine;

[0116] The analysis module 903 is used to perform transportation collaborative planning based on the operation status data and the preliminary grabbing path, and dynamically adjust the operation sequence and path planning of the bucket wheel machine by analyzing the real-time position, operation speed and load status of the transportation equipment to obtain a collaborative grabbing plan;

[0117] The prediction module 904 is used to predict the collapse area according to the distribution state information, analyze the shape and structure information of the material pile, and simulate the influence of different grabbing schemes on the shape change of the material pile to obtain the prediction result;

[0118] The optimization module 905 is used to adjust the collaborative grasping plan according to the prediction results, dynamically avoid the collapsed area with a risk level higher than the threshold by optimizing the grasping path, and generate a specific grasping action strategy to obtain a grasping operation plan.

[0119] In a specific embodiment of the present invention, the planning module 902 includes:

[0120] The first planning unit is used to perform positioning processing according to the grasping task requirements and the three-dimensional operation area model. By clustering the point cloud data of the three-dimensional operation area model, the material pile position and boundary that meet the grasping requirements are screened out, and the positioning data, spatial position and boundary data of the target grasping area are obtained;

[0121] The second planning unit is used to optimize the path starting point according to the spatial position and real-time position and attitude data, by constructing a cost function with path distance, attitude adjustment energy consumption and obstacle avoidance constraint as weights, and using the gradient descent algorithm to optimize and calculate the path starting point, and obtain the optimized path starting point parameters, which include the path starting point coordinates and attitude parameters;

[0122] The third planning unit is used to plan the local grasping path according to the boundary data and the path starting point parameters, generate path nodes by gridding on the boundary of the target grasping area, and generate the path by combining the A* algorithm with the gradient field constraint to obtain the local grasping path node sequence from the path starting point to the target grasping area;

[0123] The fourth planning unit is used to perform path smoothing processing according to the local grasping path node sequence, generate a continuous path function by interpolation, and optimize the control point position in combination with the dynamic constraint of the posture adjustment to obtain a preliminary grasping path. The preliminary grasping path includes a continuous path function and a bucket wheel machine posture adjustment instruction.

[0124] In a specific embodiment of the present invention, the analysis module 903 includes:

[0125] The first analysis unit is used to perform transportation equipment status evaluation processing according to the operation status data, calculate the idle capacity and dynamic accessibility of the transportation equipment by analyzing the real-time position, operation speed and current load status of the transportation equipment, and obtain the coordination index of the transportation equipment;

[0126] The second analysis unit is used to adjust the operation sequence according to the coordination index and the preliminary grasping path, and dynamically adjust the operation sequence of the grasping tasks by preferentially matching the idle state of the transportation equipment and the sequence constraint of the grasping path in combination with the optimal scheduling algorithm of the transportation equipment to obtain the optimized grasping sequence;

[0127] The third analysis unit is used to perform path synchronization adjustment processing according to the optimized grabbing sequence, by combining the position of the transportation equipment with the dynamic time window matching of the bucket wheel excavator grabbing path, using the path adjustment algorithm based on time window optimization, to generate a synchronous grabbing path matching the operating state of the transportation equipment, and obtain a preliminary collaborative path plan;

[0128] The fourth analysis unit is used to perform transportation collaborative optimization processing according to the preliminary collaborative path plan, and obtains a collaborative grabbing plan by dynamically adjusting the operating speed and loading rhythm of the bucket wheel excavator's grabbing path and iteratively optimizing the collaborative plan in combination with the load state constraints.

[0129] Embodiment 3:

[0130] Corresponding to the above method embodiment, an unmanned bucket wheel machine automatic control device is also provided in this embodiment. The unmanned bucket wheel machine automatic control device described below and the unmanned bucket wheel machine automatic control method described above can be referenced to each other.

[0131] Figure 3 FIG. 8 is a block diagram of an unmanned bucket wheel machine automatic control device 800 according to an exemplary embodiment. Figure 3 As shown, the unmanned bucket wheel machine automatic control device 800 may include: a processor 801, a memory 802. The unmanned bucket wheel machine automatic control device 800 may also include one or more of a multimedia component 803, an I / O interface 804, and a communication component 805.

[0132] The processor 801 is used to control the overall operation of the unmanned bucket wheel machine automatic control device 800 to complete all or part of the steps in the above-mentioned unmanned bucket wheel machine automatic control method. The memory 802 is used to store various types of data to support the operation of the unmanned bucket wheel machine automatic control device 800. These data may include, for example, instructions for any application or method operated on the unmanned bucket wheel machine automatic control device 800, and application-related data, such as contact data, sent and received messages, pictures, audio, video, etc. The memory 802 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, disk or optical disk. The multimedia component 803 may include a screen and an audio component. The screen may be, for example, a touch screen, and the audio component is used to output and / or input audio signals. For example, the audio component may include a microphone, which is used to receive external audio signals. The received audio signal may be further stored in the memory 802 or sent via the communication component 805. The audio component also includes at least one speaker for outputting audio signals. The I / O interface 804 provides an interface between the processor 801 and other interface modules, and the above-mentioned other interface modules can be keyboards, mice, buttons, etc. These buttons can be virtual buttons or physical buttons. The communication component 805 is used for wired or wireless communication between the unmanned bucket wheel machine automatic control device 800 and other devices. Wireless communication, such as Wi-Fi, Bluetooth, Near Field Communication (NFC), 2G, 3G or 4G, or a combination of one or more of them, so the corresponding communication component 805 can include: Wi-Fi module, Bluetooth module, NFC module.

[0133] In an exemplary embodiment, an unmanned bucket wheel machine automatic control device 800 can be implemented by one or more application specific integrated circuits (Application Specific Integrated Circuit, referred to as ASIC), digital signal processors (Digital Signal Processor, referred to as DSP), digital signal processing devices (Digital Signal Processing Device, referred to as DSPD), programmable logic devices (Programmable Logic Device, referred to as PLD), field programmable gate arrays (Field Programmable Gate Array, referred to as FPGA), controllers, microcontrollers, microprocessors or other electronic components, and is used to execute the above-mentioned unmanned bucket wheel machine automatic control method.

[0134] In another exemplary embodiment, a computer-readable storage medium including program instructions is also provided, and when the program instructions are executed by a processor, the steps of the above-mentioned unmanned bucket wheel machine automatic control method are implemented. For example, the computer-readable storage medium can be the above-mentioned memory 802 including program instructions, and the above-mentioned program instructions can be executed by a processor 801 of an unmanned bucket wheel machine automatic control device 800 to complete the above-mentioned unmanned bucket wheel machine automatic control method.

[0135] The above description is only a specific implementation mode of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. An automatic control method for an unmanned bucket wheel machine, characterized in that: include: Obtain the requirements of the grabbing task, the three-dimensional operation area model, the real-time position and posture data of the unmanned bucket wheel machine, the distribution status information of the material pile, and the operation status data of the transportation equipment; Performing grabbing path planning processing according to the grabbing task requirements, the three-dimensional operating area model and the real-time position and posture data, and generating a preliminary grabbing path by analyzing the operating area of ​​the bucket wheel machine, the target position of the grabbing task and the dynamic posture of the bucket wheel machine; Performing coordinated transportation planning based on the operating status data and the preliminary grabbing path, dynamically adjusting the operation sequence and path planning of the bucket wheel excavator by analyzing the real-time position, operating speed and load status of the transportation equipment, and obtaining a coordinated grabbing plan; The collapse area is predicted according to the distribution state information, and the prediction result is obtained by analyzing the morphology and structure information of the material pile and simulating the influence of different grabbing schemes on the morphology change of the material pile; The collaborative grasping scheme is adjusted according to the prediction result, the grasping path is optimized, the collapsed area with a risk level higher than a threshold is dynamically avoided, and a specific grasping action strategy is generated to obtain a grasping operation plan.

2. The automatic control method of an unmanned bucket wheel machine according to claim 1, characterized in that: According to the grabbing task requirements, the three-dimensional operating area model and the real-time position and posture data, a grabbing path planning process is performed, and a preliminary grabbing path is generated by analyzing the operating area of ​​the bucket wheel machine, the target position of the grabbing task and the dynamic posture of the bucket wheel machine, including: Perform positioning processing according to the grasping task requirements and the three-dimensional operation area model, and screen out the material pile position and boundary that meet the grasping requirements by clustering analysis on the point cloud data of the three-dimensional operation area model, and obtain the positioning data of the target grasping area, the spatial position and boundary data of the positioning data; Performing path starting point optimization processing according to the spatial position and the real-time position and attitude data, constructing a cost function with path distance, attitude adjustment energy consumption and obstacle avoidance constraint as weights, and using a gradient descent algorithm to optimize and calculate the path starting point, thereby obtaining optimized path starting point parameters, wherein the path starting point parameters include path starting point coordinates and attitude parameters; The local grasping path is planned and processed according to the boundary data and the path starting point parameters, and the path nodes are generated by gridding on the boundary of the target grasping area, and the path is generated by combining the A* algorithm with the gradient field constraint to obtain a local grasping path node sequence from the path starting point to the target grasping area; Path smoothing is performed according to the local grasping path node sequence, a continuous path function is generated by interpolation, and the control point positions are optimized in combination with the dynamic constraints of the posture adjustment to obtain a preliminary grasping path, wherein the preliminary grasping path includes a continuous path function and a bucket wheel machine posture adjustment instruction.

3. The automatic control method of an unmanned bucket wheel machine according to claim 1, characterized in that: According to the operation status data and the preliminary grabbing path, transportation collaborative planning is performed, and by analyzing the real-time position, operation speed and load status of the transportation equipment, the operation sequence and path planning of the bucket wheel excavator are dynamically adjusted to obtain a collaborative grabbing solution, including: Performing a transport equipment status evaluation process based on the operating status data, calculating the idle capacity and dynamic accessibility of the transport equipment by analyzing the real-time position, operating speed and current load status of the transport equipment, and obtaining a coordination index of the transport equipment; According to the coordination index and the preliminary grasping path, the operation sequence is adjusted, and the operation sequence of the grasping tasks is dynamically adjusted by preferentially matching the idle state of the transport equipment and the sequence constraints of the grasping path in combination with the optimal scheduling algorithm of the transport equipment to obtain an optimized grasping sequence; Performing path synchronization adjustment processing according to the optimized grabbing sequence, combining the position of the transport equipment with the dynamic time window matching of the bucket wheel excavator grabbing path, and adopting a path adjustment algorithm based on time window optimization to generate a synchronous grabbing path matching the operating state of the transport equipment, and obtain a preliminary collaborative path plan; The transport collaborative optimization process is performed according to the preliminary collaborative path plan, and the collaborative plan is iteratively optimized by dynamically adjusting the operating speed and loading rhythm of the bucket wheel excavator grabbing path and combining the load state constraints to obtain a collaborative grabbing plan.

4. The automatic control method of an unmanned bucket wheel excavator according to claim 1, characterized in that: The collapse area is predicted based on the distribution state information, and the shape and structure information of the material pile are analyzed, and the influence of different grabbing schemes on the shape change of the material pile is simulated to obtain the prediction results, including: Performing morphological feature extraction processing on the material pile according to the distribution state information, extracting surface gradient, local curvature and boundary morphological features from the three-dimensional point cloud data of the material pile using normal vector calculation and surface fitting algorithm, and marking key areas with features to obtain three-dimensional morphological feature parameters of the material pile; Performing structural stability analysis according to the three-dimensional morphological characteristic parameters, constructing a stability discrimination model based on morphological characteristics, combining gravity field analysis with contact surface distribution calculation, and conducting overall stability zoning evaluation on the material pile to obtain a stability distribution map of the material pile; Performing morphological change simulation processing according to the stability distribution diagram of the material pile, dynamically simulating different grasping schemes by using discrete element method, simulating the influence of the grasping path on the material pile structure, predicting the morphological changes caused by the grasping process, and obtaining the dynamic corresponding relationship between the grasping path and the morphological changes of the material pile; The collapse risk prediction process is performed according to the dynamic correspondence, and the high-risk collapse areas on the grasping path are marked to obtain the collapse area prediction result by calculating the risk probability distribution of each area during the grasping action.

5. The automatic control method of an unmanned bucket wheel machine according to claim 4, characterized in that: According to the three-dimensional morphological characteristic parameters, the structural stability analysis is performed. By constructing a stability discrimination model based on morphological characteristics, combined with gravity field analysis and contact surface distribution calculation, the overall stability zoning evaluation of the material pile is performed to obtain a stability distribution diagram of the material pile, including: Performing contact surface extraction processing according to the three-dimensional morphological feature parameters, analyzing the material pile point cloud data and normal vector distribution, using an Alpha shape algorithm to triangulate the material pile surface to extract contact areas where contact forces exist, and calculating the normal angle distribution and normal area weight of each contact area to obtain the contact surface distribution characteristics of the material pile; The gravity distribution is calculated and processed according to the contact surface distribution characteristics, by dividing the three-dimensional shape of the material pile into finite volume units, using the finite element method to calculate the center of gravity coordinates and gravity distribution of each unit, and combining the normal characteristics of the contact surface to perform mechanical decomposition of the support force and sliding force of each unit, so as to obtain the gravity field distribution inside the material pile; Performing stability analysis according to the gravity field distribution, calculating the anti-slip force of each area on the contact surface based on a preset Coulomb friction model, calculating the stability factor by combining the gravity field distribution and the support force of the local contact point, and classifying the stability of each contact surface area according to the critical value to obtain the local distribution of the stability factor; According to the local distribution of the stability factor, zoning is performed. By constructing a global assessment model based on the stability factor, the locally distributed stability factor is weighted interpolated and combined with the transition continuity between regions, the high-risk areas are marked and the risk level is classified to obtain the overall stability distribution map of the material pile.

6. The automatic control method of an unmanned bucket wheel machine according to claim 4, characterized in that: According to the stability distribution diagram of the material pile, the morphological change simulation is performed, and different grasping schemes are dynamically simulated by using the discrete element method to simulate the influence of the grasping path on the material pile structure, predict the morphological changes caused by the grasping process, and obtain the dynamic corresponding relationship between the grasping path and the morphological changes of the material pile, including: A grasping action mechanics model is constructed based on the stability distribution diagram of the material pile. The material pile is decomposed into a number of discrete particle units by combining the stability distribution diagram of the material pile with the geometric model of the grasping tool, and a particle contact mechanics algorithm based on the discrete element method is used. The grasping action mechanics model of the material pile is constructed by combining the calculation of the contact points, contact forces and friction forces between the particles, and the initial state parameters describing the internal stress of the material pile and the movement of the particles are obtained. According to the initial state parameters, dynamic simulation processing of particle movement is performed, and by analyzing the particle force at each path point of the grasping path, the force balance state and movement trajectory of the particles under the action of the grasping tool are calculated step by step and iteratively, and the contact relationship and stacking form between the particles are dynamically updated to obtain the dynamic morphological change of the grasping path on the distribution of the particles in the material pile; Extracting the affected area of ​​the morphological change according to the dynamic morphological change, and performing spatial clustering analysis on the changed areas of particle displacement and stacking morphology in the simulation results, identifying the high displacement area, the slip boundary area and the local collapse area within the action range of the grasping path, and obtaining the affected area of ​​the morphological change; A dynamic correspondence is established according to the impact area of ​​the morphological change. By matching and analyzing the dynamic characteristic parameters of the impact area with the time series of the grasping path, an association modeling method based on time series interpolation and regional influence intensity weighting is used to generate a grasping dynamic correspondence diagram, and the dynamic correspondence between the grasping path and the morphological change of the material pile is obtained.

7. The automatic control method of an unmanned bucket wheel machine according to claim 1, characterized in that: The collaborative grasping scheme is adjusted according to the prediction results, the grasping path is optimized, the collapsed area with a risk level higher than the threshold is dynamically avoided, and a specific grasping action strategy is generated to obtain an optimized grasping operation scheme, including: According to the prediction results and the collaborative grasping scheme, high-risk areas are dynamically identified and processed, and the marked high-risk areas of collapse are spatially overlapped, and the intersection nodes of the grasping path and the collapsed area are located using the KD tree, and the areas to be avoided are screened according to the risk level threshold, so as to obtain the conflict node set between the grasping path and the high-risk area; Performing path avoidance processing according to the conflict node set, by regenerating an alternative path from the grabbing starting point to the target area in the three-dimensional working area model, defining constraint conditions as avoiding conflict nodes, shortest path priority, and bucket wheel machine posture adjustment restrictions, to obtain an optimized grabbing path; The grasping action strategy is optimized according to the optimized grasping path. By sampling each path point and combining the real-time position and posture data of the bucket wheel machine, a motion parameter optimization algorithm based on gradient descent is used to calculate the grasping depth, angle adjustment and force control of the grasping action, and the action strategy corresponding to each position on the grasping path is obtained; An operation plan is generated based on the grasping action strategy and the optimized grasping path. By integrating the dynamic avoidance path with the grasping action strategy, the path is continuously optimized using cubic B-spline curve interpolation. At the same time, the rhythm of the grasping action is synchronously adjusted in combination with the operating status of the transportation equipment to obtain an optimized grasping operation plan.

8. An automatic control system for an unmanned bucket wheel excavator, characterized in that: include: The acquisition module is used to obtain the grasping task requirements, the three-dimensional operation area model, the real-time position and posture data of the unmanned bucket wheel machine, the distribution status information of the material pile, and the operation status data of the transportation equipment; A planning module, used for performing grabbing path planning processing according to the grabbing task requirements, the three-dimensional operating area model and the real-time position and posture data, and generating a preliminary grabbing path by analyzing the operating area of ​​the bucket wheel machine, the target position of the grabbing task and the dynamic posture of the bucket wheel machine; An analysis module is used to perform transportation collaborative planning according to the operation status data and the preliminary grabbing path, and dynamically adjust the operation sequence and path planning of the bucket wheel excavator by analyzing the real-time position, operation speed and load status of the transportation equipment to obtain a collaborative grabbing plan; A prediction module, used to predict the collapse area according to the distribution state information, analyze the morphology and structure information of the material pile, and simulate the influence of different grabbing schemes on the morphology change of the material pile to obtain the prediction result; The optimization module is used to adjust the collaborative grasping scheme according to the prediction results, dynamically avoid the collapsed area with a risk level higher than a threshold by optimizing the grasping path, and generate a specific grasping action strategy to obtain a grasping operation plan.

9. The automatic control system of an unmanned bucket wheel excavator according to claim 8, characterized in that: The planning module includes: A first planning unit is used to perform positioning processing according to the grasping task requirements and the three-dimensional operation area model, and to screen out the material pile position and boundary that meet the grasping requirements by clustering analysis on the point cloud data of the three-dimensional operation area model, so as to obtain the positioning data of the target grasping area, the spatial position and boundary data of the positioning data; A second planning unit is used to optimize the path starting point according to the spatial position and the real-time position and attitude data, by constructing a cost function with path distance, attitude adjustment energy consumption and obstacle avoidance constraint as weights, and using a gradient descent algorithm to optimize and calculate the path starting point, so as to obtain optimized path starting point parameters, wherein the path starting point parameters include path starting point coordinates and attitude parameters; A third planning unit is used to plan a local grasping path according to the boundary data and the path starting point parameters, generate path nodes by gridding on the boundary of the target grasping area, and generate a path by combining the A* algorithm with the gradient field constraint to obtain a local grasping path node sequence from the path starting point to the target grasping area; The fourth planning unit is used to perform path smoothing processing according to the local grasping path node sequence, generate a continuous path function by interpolation, and optimize the control point position in combination with the dynamic constraint of the posture adjustment to obtain a preliminary grasping path, wherein the preliminary grasping path includes a continuous path function and a bucket wheel machine posture adjustment instruction.

10. The automatic control system of an unmanned bucket wheel excavator according to claim 8, characterized in that: The analysis module comprises: A first analysis unit is used to perform a transportation equipment status evaluation process according to the operation status data, calculate the idle capacity and dynamic accessibility of the transportation equipment by analyzing the real-time position, operation speed and current load status of the transportation equipment, and obtain a coordination index of the transportation equipment; The second analysis unit is used to adjust the operation sequence according to the coordination index and the preliminary grasping path, and dynamically adjust the operation sequence of the grasping tasks by preferentially matching the idle state of the transportation equipment and the sequence constraint of the grasping path in combination with the optimal scheduling algorithm of the transportation equipment to obtain an optimized grasping sequence; The third analysis unit is used to perform path synchronization adjustment processing according to the optimized grabbing sequence, by combining the position of the transportation equipment with the dynamic time window matching of the bucket wheel excavator grabbing path, using the path adjustment algorithm based on time window optimization, to generate a synchronous grabbing path matching the operating state of the transportation equipment, and obtain a preliminary collaborative path plan; The fourth analysis unit is used to perform transportation collaborative optimization processing according to the preliminary collaborative path plan, dynamically adjust the operating speed and loading rhythm of the bucket wheel excavator's grabbing path, iteratively optimize the collaborative plan in combination with the load state constraints, and obtain a collaborative grabbing plan.

Images