Intelligent decision-making task scheduling method for steel grabbing machine in scrap steel yard

Abstract

The invention relates to an intelligent decision task scheduling method for a steel grabbing machine in a scrap steel stock yard, and belongs to the technical field of steel grabbing machine scheduling, and the method comprises the following steps: S1, carrying out scene initialization according to input processed gridding data and material pile information, and generating a scene structural body; s2, updating grid data of material piles and truck warehouses in the scene, and selecting a grabbing point and a placing point of the round; s3, the steel grabbing machine carries out automatic operation according to the grabbing point and the placing point selected in the round; and S4, the steps S2-S3 are repeated till the truck warehouse is full, and steel grabbing is finished. The operation efficiency, the grabbing stability and the automation degree of the steel grabbing machine are remarkably improved, and the technical effect higher than that of a manual processing mode in the prior art is achieved.

Description

Technical Field

[0001] The invention belongs to the technical field of steel grabber scheduling, and relates to an intelligent decision-making task scheduling method for steel grabbers in a scrap steel yard. Background Art

[0002] In traditional scrap steel yard operations, steel grabbers are typically manually controlled. However, the complexity of the material stacks often makes manual operation reliant on experience. This approach is not only inefficient but also suffers from low grasping accuracy. Therefore, the project's goal is to achieve unmanned operation of the steel grabber through an intelligent grasping decision-making algorithm. In the profile scenario, the unmanned grasping system must analyze factors such as the geometric characteristics of the material stack, the dynamic parameters of the steel grabber, and the volume of the truck cargo hold to make real-time, accurate grasping and placement decisions.

[0003] The main technical problems encountered in the decision-making process of steel grabbers in profile scenarios can be summarized as follows:

[0004] (1) Difficulty in selecting grabbing points and placement points: In the scenario of profile stacking, different profiles have different shapes, sizes, and placement positions, which makes it difficult to accurately select the appropriate grabbing points. The selection of grabbing points must comprehensively consider the operability of the steel grabber, the stability of the profile, and the efficiency of grabbing. Similarly, the selection of placement points also faces similar problems. The placement points must not only meet the size and space restrictions of the profiles, but also ensure stability during the placement process to prevent the material from falling;

[0005] (2) Grasping stability and efficiency: Stability is particularly critical. During the grasping process, it is necessary to ensure that the profile does not fall out of the steel claw. This requires judging the stability of the grasping window, calculating the center of gravity distribution of the profile, and the force and angle of the steel claw's grasping. In terms of efficiency, it is necessary to maximize the amount of material grasped each time, while reducing the time for grasping and placing, thereby improving overall work efficiency;

[0006] (3) Complex scene initialization and data processing: During the operation of the steel grabber, a large amount of dynamic data about the profile stack, the truck cargo hold, and the steel grabber itself must be processed. This data includes the geometric information of the profile, mesh data, the remaining space on the truck, etc. The initialization process is complex and requires high data processing capabilities;

[0007] (4) Intelligent decision-making for unmanned operation: The unmanned grasping process requires real-time decisions on grasping points and placement points through intelligent algorithms, taking into account multiple variables such as grasping angle, depth, quantity and location of grasped materials, etc. These decision-making algorithms need to ensure that the steel grasper can continue to work efficiently during multiple operations. Summary of the Invention

[0008] In view of this, the object of the present invention is to provide an intelligent decision-making task scheduling method for a steel grabber in a scrap steel yard.

[0009] In order to achieve the above object, the present invention provides the following technical solutions:

[0010] An intelligent decision-making task scheduling method for a steel grabber in a scrap steel yard comprises the following steps:

[0011] S1: First, the scene is initialized based on the input processed grid data and stockpile information to generate the scene structure;

[0012] S2: Update the mesh data of the pile and truck cargo hold in the scene, and select the grab and placement points for this round;

[0013] S3: The steel grabber automatically operates according to the grabbing and placing points selected in this round;

[0014] S4: Repeat steps S2-S3 until the truck cargo hold is full and the steel grabbing process ends.

[0015] Furthermore, the scene initialization includes the following steps:

[0016] Input the material yard information, material pile grid data, truck cargo hold grid data, basic static and dynamic information of the steel grabber, basic truck information, obstacle information, and profile information;

[0017] The input data is stored in a scene structure; the scene structure contains all the elements of the scene and integrates this information to form a complete scene that can be used.

[0018] Next, determine whether the truck is full by following these steps:

[0019] Input scene structure, which contains the truck cargo area and related scene information;

[0020] Traverse the grid data of the warehouse, check the information in each grid, and determine whether it exceeds the height limit of the truck warehouse. If it exceeds, it means that the truck warehouse is full, otherwise it is not full.

[0021] Furthermore, the updating of the grid data of the stockpile and the truck cargo hold in the scene in step S2 specifically includes the following steps:

[0022] Input the grabbing information of the steel grabber after each grab, the information of the truck cargo hold, the mesh data of the stockpile, the mesh data of the truck cargo hold, and the profile information in the scene;

[0023] Obtain updated mesh data and the results of steel grabbing by the grabber;

[0024] Update the scene structure and apply the new data to the scene;

[0025] Output the updated scene structure.

[0026] Furthermore, the placement point selection process is as follows:

[0027] Input scene structure, which contains the information about the truck warehouse and material pile in the current scene;

[0028] Determine whether this is the first time a placement point is selected. If so, select a placement point in front of the truck cargo hold, where the grid height does not exceed the height limit of the truck cargo hold. If this is not the first time, determine whether the height of the previous placement point has exceeded the height of the truck cargo hold. If not, continue to use the previous placement point. If the height has exceeded, select a new placement point to ensure that the height limit of the truck cargo hold is not exceeded.

[0029] Output the placement point position selected in this round.

[0030] Furthermore, the grab point selection process is as follows:

[0031] Input scene structure;

[0032] Determine whether the current grabbing window can be closed or ended. If so, calculate the grabbing position and determine the grabbing amount based on the material in the window. Calculate the time and path required to reach the grabbing position and determine whether the grabbing window is full. If the grabbing window is not full, calculate the grabbing percentage and related information.

[0033] After determining whether all the capture windows are full or invalid, confirm whether it is necessary to switch the capture window, and finally output the result of the capture operation.

[0034] Furthermore, the grab point selection specifically includes the following steps:

[0035] Step 1: Feasibility determination of the capture window: If feasible, go to step 2; otherwise, go to step 7. The feasibility determination of the window refers to:

[0036] If the unit grid height variance is less than the preset threshold, that is, the surface of the entire pile is relatively flat, it is considered infeasible;

[0037] If the height of the pile of molding materials from the ground is less than the preset threshold, making it difficult to grab, it is also considered unfeasible;

[0038] Step 2: Calculate the grabbing position: First, determine the selection position: select the center of the rectangular area, calculate the ridge line at the center to determine the fine-grained orientation of the steel claw, and use the center of the ridge line as the grabbing position. The grabbing depth is based on the highest unit grid height. At the same time, the opening size is determined according to the shearing machine capacity limit.

[0039] Step 3: Calculate the estimated weight of the profile within the window: Calculate the estimated profile volume as the weight based on the average height of the unit grid within the grasping range, the grasping depth and the size of the claw opening;

[0040] Step 4: Calculate the Euclidean distance and time required for the claw to move to the grasping point;

[0041] Step 5: Calculate the cost-effectiveness of capturing in this window. The calculation formula is as follows:

[0042] R=W / T

[0043] Where T represents the time required for the steel claw to move to the grasping position, W represents the estimated weight of the steel, and R represents the cost-effectiveness, that is, the greater the weight of steel grasped per unit time, the better;

[0044] Step 6: Determine whether all windows have been calculated; if so, go to step 8; otherwise, go to step 7;

[0045] Step 7: Window switching: Slide the window at a fixed step size and go to step 2;

[0046] Step 8. Select the best cost-effective window: Based on the cost-effectiveness, select the best window corresponding to the steel grabbing position, depth, and steel claw direction.

[0047] The beneficial effects of this invention are that, through intelligent algorithms, this solution overcomes the shortcomings of existing technologies, such as inaccurate grasping and placement points, low efficiency, poor grasping stability, and the inability to achieve unmanned operation. Through real-time data updates, intelligent path planning, and cost-effective optimization, this solution significantly improves the efficiency, grasping stability, and automation of the steel grabber, achieving technical results superior to existing technologies and manual processing methods.

[0048] Other advantages, objects, and features of the present invention will be described in part in the following description and, in part, will be apparent to those skilled in the art upon examination of the following description or may be learned from practice of the present invention. The objects and other advantages of the present invention may be realized and obtained through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] In order to make the purpose, technical solutions and advantages of the present invention more clear, the present invention will be described in detail below with reference to the accompanying drawings, in which:

[0050] Figure 1 This is the overall flow chart of the intelligent decision-making task scheduling method for steel grabbers in scrap steel yards;

[0051] Figure 2 Initialize the process for the scene;

[0052] Figure 3Flowchart for determining truck status;

[0053] Figure 4 Updated maps for stockpile and truck bed data;

[0054] Figure 5 Select the flow chart for the placement point;

[0055] Figure 6 Select a flow chart for the grab point;

[0056] Figure 7 Flowchart for judging whether the sub-window meets the crawling requirements;

[0057] Figure 8 Flowchart for calculating the grab position within the subwindow;

[0058] Figure 9 Flowchart for calculating the weight of steel to be grabbed in the sub-window;

[0059] Figure 10 Flowchart for calculating the time T required for the steel claw to reach the steel grabbing position;

[0060] Figure 11 Flowchart for calculating the reward value for steel grabbing within a feasible sub-window;

[0061] Figure 12 Flowchart for selecting the best grabbing sub-window;

[0062] Figure 13 Generate a flow chart for the results;

[0063] Figure 14 To determine whether the steel grabber has completed the round of grabbing action flow chart;

[0064] Figure 15 This is a visualization diagram of the simulation results;

[0065] Figure 16 is the initial material state diagram;

[0066] Figure 17 To grab the material status once;

[0067] Figure 18 This is the state diagram of the material grabbing completed;

[0068] Figure 19 Output graph for simulation results. DETAILED DESCRIPTION

[0069] The following describes the embodiments of the present invention by means of specific examples, and those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present invention, and the following embodiments and features in the embodiments can be combined with each other without conflict.

[0070] It should be noted that the illustrations provided in the following embodiments are merely schematic illustrations of the basic concept of the present invention. Therefore, the illustrations only show components related to the present invention and are not drawn according to the number, shape, and size of components in actual implementation. In actual implementation, the type, quantity, and proportion of each component may be changed arbitrarily, and the component layout may also be more complex.

[0071] In the following description, numerous details are discussed to provide a more thorough explanation of the embodiments of the present invention. However, it will be apparent to those skilled in the art that the embodiments of the present invention may be practiced without these specific details. In other embodiments, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring the embodiments of the present invention.

[0072] Example 1:

[0073] This invention provides an intelligent decision-making and task scheduling method for scrap steel yard grabbers. By introducing intelligent grab and placement point selection algorithms, stability and efficiency optimization, real-time scene data updates, and an intelligent decision-making mechanism for unmanned operation, this method effectively addresses the technical challenges of grabbers in profile scenarios. This technical solution ensures the efficiency and stability of each grab, while also improving overall operational efficiency through intelligent path planning and cost-effectiveness calculation.

[0074] The system first initializes the scenes of the material yard, steel grabber and truck based on the input grid data. Including geometric information of the profile, working parameters of the steel grabber, remaining space in the truck cargo hold and other information, the system will organize this information into a grid data structure to support subsequent grabbing and placement operations. The grid data is used to represent the geometric structure of stacked materials, truck cargo holds and other scenes. Each grid contains position information (such as x, y, z coordinates) and height information to assist the steel grabber in making operational decisions. During the grabbing process, the grid data after each grab will be updated. These updated data will be used to determine whether the truck is full and for subsequent grabbing and placement point selection. The system will also adjust the grabbing strategy based on the dynamic information of the steel grabber (such as grabbing angle, depth and opening size).

[0075] In a real-world scenario, once a truck arrives near a steel grabber, the truck driver can manually bind the truck to the service platform, notifying it of new tasks. The steel grabber can then autonomously allocate tasks. Ultimately, this allows the unmanned steel grabber to operate autonomously even in the event of a failure in the operations control center or communication interruption.

[0076] To enable unmanned operation of the steel grabber, the algorithm first intelligently generates a set of candidate grab and placement points. By analyzing the shape of the scrap pile and the dynamic parameters of the steel grabber, the algorithm determines the feasibility of each candidate grab point and calculates its cost-effectiveness, ultimately selecting the grab point with the highest cost-effectiveness.

[0077] The algorithm calculates the optimal path for the grabber to move from its current position to the target grasping point, taking into account the grabber's dynamic constraints and obstacle information. Dynamic constraints refer to the physical constraints imposed on the grabber's movement, such as the grabber's maximum reach, rotation angle, and lift height. These constraints influence the grabber's motion path planning. This ensures that the grabber avoids collisions with obstacles during the grasping process and minimizes path travel time.

[0078] The selection of grasping points is based on the calculation of the profile geometry, including geometric data such as the profile's center position, orientation, length, and height. Through real-time scene information (such as the grid data of the profile pile and the truck's cargo hold data), the system can dynamically generate a set of candidate grasping points. The system will judge the operability of the grasping based on the height distribution of the unit grid and the feasibility of the grasping window. For example, the grasping position is determined by calculating the ridge line (the line of the highest point on the profile surface), and the opening angle and grasping depth of the steel claw are determined according to the capacity limit of the shearing machine.

[0079] The placement points are selected according to specific rules, such as prioritizing placement points from the rear of the truck bed forward and ensuring that the steel grabber's gripping operation does not exceed the height limit of the truck bed. For the placement of profiles, the algorithm intelligently generates a set of feasible placement points based on the truck's current position, available cargo space, and profile size.

[0080] To address the potential for profiles to fall during the grasping process, the stability of each grasping window is calculated by analyzing the height distribution of each unit grid within the profile pile. The grasping window refers to the graspable area defined by the steel gripper during each grasping operation. The size of the grasping window is dynamically adjusted based on the degree of opening of the steel claws and the distribution of the material. The grasping algorithm calculates the center of gravity of the profile to ensure that the center of gravity is concentrated and that instability due to eccentricity is not caused during grasping. At the same time, the algorithm determines the safety and effectiveness of the grasping process based on the distribution of objects within the grasping window. The selection of grasping points must not only consider feasibility, but also calculate the efficiency of grasping through an optimization algorithm. By calculating the cost-effectiveness of grasping R = W / T, that is, the ratio of the weight W of the profile grasped per unit time to the time T required for grasping, the system can ensure maximum efficiency in each grasping. At the same time, the motion path planning algorithm of the steel gripper can ensure the shortest path between the grasping and placement points, thereby reducing ineffective movement time.

[0081] There are three types of steel material piles in the scrap steel yard, including cylindrical ones. Currently, traditional manually operated steel grabbers grab steel and feed it to the shearing machine. Considering the transformation of the steel grabber to realize unmanned grabbing and feeding of steel and optimize efficiency, the algorithm needs to give the specific grabbing position of the steel grabber each time. This position must take into account the feasibility of grabbing steel, including the current maximum capacity limit of the shearing machine and the accuracy and stability of the grabbing quantity, and must also take into account the total time efficiency of completing all feeding tasks. That is, try to make the steel grabber obtain enough steel in each grab and make the grabbing process take the least time.

[0082] Through the implementation of the above technical solutions, the steel grabber has achieved highly intelligent selection of gripping and placement points in the profile scenario. It can automatically select the optimal operating position based on the geometric characteristics and dynamic data of the profile, significantly improving the stability of the gripping process and reducing the risk of material falling. In addition, gripping efficiency is maximized by optimizing the cost-effectiveness algorithm, ensuring that more material is gripped per unit time. At the same time, the system also realizes unmanned operation. The steel grabber can automatically complete the gripping and placement tasks without human intervention. Combined with intelligent path planning, it effectively shortens the operation time. Through real-time data updates and feedback mechanisms, it continuously optimizes subsequent operations to ensure the efficiency, accuracy and safety of the overall operation.

[0083] Example 2:

[0084] In this embodiment, if Figure 1As shown, a method for intelligent decision-making and task scheduling for a scrap steel yard grabber includes the following steps: First, the scene is initialized based on the processed grid data, the stockpile, and other information. Based on the grid data of the truck cargo hold, the algorithm determines whether the truck is full. If so, the algorithm outputs the result; otherwise, the grabber enters a selection loop for selecting grab points and placement points. The algorithm updates the grid data of the stockpile and truck cargo hold within the scene, selects grab points, selects placement points, and sends the selection results to the grabber for operation. The system then determines whether the current grabbing task is complete. If so, the system enters the next round of grab point and placement point selection. Otherwise, the algorithm waits and determines again.

[0085] like Figure 2 The figure shows the process of initializing a scene. The input includes material yard information, material pile mesh data, truck cargo hold mesh data, basic static and dynamic information about the steel grabber, basic truck information, obstacle information, and material profile information. During initialization, the program reads this input data and stores it in a scene structure, ultimately generating the scene structure. This structure contains all the elements of the scene and integrates this information to form a complete, usable scene. The entire process, from data reading to scene generation, is automated, ensuring that all input data is processed to form a scene suitable for further operation or simulation.

[0086] like Figure 3 The figure shows the process of determining whether the truck is full based on the warehouse grid data. The process of updating the pile data is as follows Figure 5 As shown, the grid data is processed, and the grid information within the operating range is filtered and saved according to the operating range of the steel grabber, and the rest of the information is discarded.

[0087] like Figure 4 The figure shows the process of updating the mesh data for the stockpile and truck cargo area in the scene. First, the mesh data is input for processing. Based on the steel grabber's operable range, the mesh information within the operating range is filtered and saved. The remaining information is discarded, and finally, the updated mesh data is output. This process effectively updates the stockpile, truck, and other scene data in real time, ensuring the accuracy and real-time nature of the scene information.

[0088] like Figure 5As shown, the capture of placement points specifically includes: first, the input is the scene data g_tScene, which contains relevant information such as the truck cargo hold and material pile in the current scene. The system first determines whether this is the first time a placement point is selected. If so, it selects a placement point from the corresponding front of the truck, and the grid height of the placement point does not exceed the height limit of the truck cargo hold. If it is not the first time, it determines whether the height of the last placement point has exceeded the height of the truck cargo hold. If not, the system will continue to use the last placement point. If the height has exceeded, a new placement point will be selected to ensure that the height limit of the truck cargo hold is not exceeded. Finally, the output result is tPlacePoint, which is the placement point position selected by the system based on the conditions. Through this logic, it can ensure that the appropriate placement position is selected during the loading process, so that the entire loading process proceeds smoothly.

[0089] like Figure 6 As shown, this is the process of selecting grab points and determining the information required for grabbing. First, the system inputs the scene data g_tScene. Then, it determines whether the current grab window can be closed or ended. If so, the grab position is calculated, and the grab amount is determined based on the material in the window. At the same time, the system calculates the time and path required to reach the grab position, and determines whether the grab window is full. If the grab window is not full, the grab percentage and related information are calculated. After determining whether all grab windows are full or invalid, the system will further confirm whether there is a need to switch the grab window, and finally the system outputs vGrabResult or tGrabResult, which is the result of the grab operation. Through this process, the system can intelligently select grab points and perform grab operations based on the actual situation in the scene to ensure the effectiveness and accuracy of the grab process. The basic process of grab point selection is as follows:

[0090] Step 1: Determine the feasibility of the gripping window. If feasible, proceed to Step 2; otherwise, proceed to Step 7. The feasibility of the window depends on whether the gripper can easily engage, which is primarily affected by the height distribution of the pile. A small unit grid height variance, meaning the entire pile surface is relatively flat, is considered unfeasible. Furthermore, if the pile is too close to the ground, gripping becomes difficult and is also considered unfeasible.

[0091] Step 2: Calculate the grabbing position. First, determine the location: select the center of the rectangular area, calculate the ridgeline at the center to determine the fine-grained orientation of the steel claw, and use the center of the ridgeline as the grabbing position. The grabbing depth is based on the highest unit grid height. At the same time, the opening size is determined according to the shearing machine capacity limit.

[0092] Step 3: Calculate the estimated weight of the material within the window. The estimated volume of the material is calculated based on the average height of the grid cells within the grasping range, the grasping depth, and the size of the claw opening.

[0093] Step 4: Calculate the Euclidean distance and time required for the claw to move to the grasping point.

[0094] Step 5: Calculate the cost-effectiveness of the grabbing process within this window. R = W / T, where T represents the time it takes for the claw to move to the grabbing position, W represents the estimated weight of the steel, and R represents the cost-effectiveness. This means the greater the weight of steel grabbed per unit time, the better.

[0095] Step 6: Determine whether all windows have been calculated. If so, go to step 8; otherwise, go to step 7.

[0096] Step 7: Window switching. Slide the window with a fixed step size and go to step 2.

[0097] Step 8: Select the best cost-effective window. Based on the cost-effectiveness, select the best window corresponding to the steel grabbing position, depth, and claw orientation.

[0098] like Figure 7 The figure shows the initial process of determining whether a subwindow meets the gripper's lower jaw requirements in the first step. First, vStockPileGrid (stockpile grid data) and tSubWindowInfo (subwindow information) are input. Next, the system calculates the average height, maximum height, and height variance of each cell grid within the subwindow. These calculations form the basis for determining whether the gripper can grasp the area. The system then determines whether these calculated values meet the lower jaw height and angle requirements to determine whether to proceed. Finally, a Boolean value, bGrabRequirement, is output, indicating whether the grasping requirements are met. The innovation of this technical solution lies in its multi-level analysis of cell grid heights, considering not only the average height but also the maximum height and height variance. This approach effectively avoids grasping failures or damage to the gripper when grasping on irregular surfaces. This multi-dimensional data analysis enables the system to more intelligently determine the feasibility of the grasping area, ensuring safe and accurate grasping. Furthermore, the calculation of height variance enables the gripper to operate precisely in areas with large height variations, improving overall grasping efficiency and stability. This innovative solution is particularly suitable for complex and dynamically changing material yard environments, and improves the adaptability of the steel grabber.

[0099] like Figure 8The figure shows the process of calculating the grasping position within the subwindow, described in step 2. First, the input is tSubWindowInfo (subwindow information). The system first calculates the highest point within the subwindow, the so-called "ridgeline." Next, the system calculates the optimal grasping position based on the height of the "ridgeline." After determining the grasping position, the system calculates the gripper opening size based on the volume of the material being grasped and the specific conditions of the grasping area. The system then further calculates the grasping depth to ensure effective and safe grasping. Finally, tGrapProfilesInfo is output, which contains updated information such as the grasping depth, gripper opening size, and grasping direction. This solution determines the grasping point location through multi-dimensional calculations. First, the precise location of the "ridgeline" is a key innovation. By identifying the highest point, the system can effectively avoid unnecessary empty grasps or overgrasping of material, improving grasping accuracy and efficiency. Secondly, the grasping opening size and grasping depth are dynamically adjusted based on the actual material characteristics and grasping task requirements. This adaptive design ensures flexible and reliable grasping operations. Especially in complex material pile scenarios, this multi-level adjustment can significantly improve grasping efficiency and avoid grasping failure or equipment damage caused by openings that are too large or too shallow.

[0100] like Figure 9 The figure shows the process of calculating the weight of the steel material to be grasped within the subwindow, described in step 3. First, vGrapProfiles is input, which is the grid information near the grasping location. The system estimates the volume of the steel material within this area based on the height information of the grids near the grasping location. Based on this volume information, the system further calculates the total weight of the steel material. Finally, dTotalVolume is output, which is the total weight of the steel material grasped within the subwindow. The innovation of this solution lies in its comprehensive analysis of the grid heights around the grasping location. By reading the grid heights near the grasping location, the system can accurately estimate the volume of the steel material within this area. This estimation method based on grid height differences is highly innovative and avoids the inaccuracies associated with traditional methods that rely directly on preset data or average values. This method of calculating volume based on actual grid heights can significantly improve the accuracy of weight estimation, especially in cases of irregular stockpiles and complex steel shapes. Furthermore, by dynamically adjusting the volume calculation near the grasping point in real time, this solution ensures that the system can adapt to different scene changes and material stacking configurations, thereby improving the accuracy and efficiency of the entire grasping process.

[0101] like Figure 10The figure shows the fourth step, which calculates the time T required for the gripper to reach the steel grabbing position within the feasible subwindow. First, the input is tGrapProfilesInfo, which is the steel grabbing information within the subwindow. The calculation mainly requires the coordinates of the gripper's current position and the grabbing position. Based on the Euclidean distance between the two and the gripper's average movement speed, the time T required for the gripper to reach the steel grabbing position within the feasible subwindow is calculated.

[0102] Figure 11 This article describes the process of calculating the reward for grabbing steel within a feasible subwindow. First, the input is tGrapProfilesInfo, which is the information about the steel grab within the subwindow. Based on this information, the system calculates the reward value R using the formula R = W / T, where W represents the weight of the steel grabbed and T represents the time spent grabbing. This calculation allows the system to evaluate the efficiency of the grab operation. Finally, the updated tGrapProfilesInfo is output, which is the grab information that includes the grab reward value.

[0103] Figure 12 To select the optimal grab window, the flow chart first inputs vAllGrapProfilesInfo, which represents the steel grab profile for all possible grab subwindows. The system compares the reward value R for each subwindow and selects the subwindow with the highest reward value as the final grab window. Finally, the output is vGrabResult, which contains detailed information about the grab operation.

[0104] Figure 13 This describes the process of sending the pick and place points generated in the current round to the steel grabber according to the communication protocol. First, the input is tResult, which represents the pick and place point selection results for this round. The system then sends this pick and place point information to the steel grabber according to the preset communication protocol for operation. Finally, the system outputs communication data, indicating that the pick and place points have been successfully sent through the communication protocol.

[0105] Figure 14 This process uses a communication protocol to determine whether the grabber has completed its current grabbing operation. First, the input is tTruck, which represents the truck information after the grabbing operation. The system then obtains the completion status of the grabber through the communication protocol and returns the result. Finally, it outputs a Boolean value (bool) indicating whether the grabber has completed the grabbing operation.

[0106] When conducting simulation tests, we use the three-dimensional coordinate data of the real material pile as the test data, converting the profile material into raster data. The simulation results are visualized as shown in the figure below. Figure 15 As shown, the effect of continuous grabbing is as follows Figure 16-18As shown. According to the simulation results, the amount of material grasped and the grasping center can be output in real time. The results are shown in Figure 19 shown.

[0107] Example 3:

[0108] An electronic device comprising a memory and a processor;

[0109] The memory is used to store computer programs;

[0110] The processor is configured to implement the method described in Example 1 when executing the computer program.

[0111] Example 4:

[0112] A computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method described in Example 1 is implemented.

[0113] Example 5:

[0114] A computer program product includes a computer program, which implements the method described in embodiment 1 when executed by a processor.

[0115] In the above embodiments, references to "this embodiment" in the specification indicate that a particular feature, structure, or characteristic described in conjunction with the embodiment is included in at least some embodiments, but not necessarily all embodiments. Multiple occurrences of "this embodiment" do not necessarily refer to the same embodiment.

[0116] In the above embodiments, references to "this embodiment" in the specification indicate that a particular feature, structure, or characteristic described in conjunction with the embodiment is included in at least some embodiments, but not necessarily all embodiments. Multiple occurrences of "this embodiment" do not necessarily refer to the same embodiment.

[0117] In the above embodiments, although the invention has been described in conjunction with specific embodiments thereof, many alternatives, modifications, and variations of these embodiments will be apparent to those skilled in the art based on the foregoing description. For example, other memory structures (e.g., dynamic RAM (DRAM)) may be used with the embodiments discussed. The embodiments of the present invention are intended to encompass all such alternatives, modifications, and variations that fall within the broad scope of the appended claims.

[0118] Regarding the computer-readable storage medium in this embodiment, those skilled in the art will appreciate that all or part of the steps in the aforementioned method embodiments can be implemented using hardware associated with the computer program. The aforementioned computer program can be stored in a computer-readable storage medium. When executed, the program performs the steps in the aforementioned method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

[0119] The electronic terminal provided in this embodiment includes a processor, a memory, a transceiver and a communication interface. The memory and the communication interface are connected to the processor and the transceiver and complete communication with each other. The memory is used to store computer programs, the communication interface is used for communication, and the processor and the transceiver are used to run computer programs so that the electronic terminal executes the various steps of the above method.

[0120] In this embodiment, the memory may include a random access memory (RAM), and may also include a non-volatile memory (non-volatile memory), such as at least one disk storage.

[0121] The above-mentioned processor can be a general-purpose processor, including a central processing unit (CPU), a network processor (NP), etc.; it can also be a digital signal processor (DSP), an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, and discrete hardware components.

[0122] The present invention can be used in a wide variety of general-purpose or special-purpose computing system environments or configurations, such as personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments that include any of the above.

[0123] The present invention may be described in the general context of computer-executable instructions, such as program modules, executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, and the like that perform specific tasks or implement specific abstract data types. The present invention may also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communications network. In a distributed computing environment, program modules may be located in both local and remote computer storage media, including storage devices.

[0124] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not limiting. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention can be modified or replaced by equivalents without departing from the purpose and scope of the technical solutions, which should all be included in the scope of the claims of the present invention.

Claims

1. An intelligent decision-making task scheduling method for a steel grabber in a scrap steel yard, characterized by: The following steps are involved: S1: First, the scene is initialized based on the input processed grid data and stockpile information to generate the scene structure; S2: Update the mesh data of the pile and truck cargo hold in the scene, and select the grab and placement points for this round; S3: The steel grabber automatically operates according to the grabbing and placing points selected in this round; S4: Repeat steps S2-S3 until the truck cargo hold is full and the steel grabbing process ends.

2. The intelligent decision-making task scheduling method for scrap steel yard grabbers according to claim 1 is characterized by: The scene initialization includes the following steps: Input the material yard information, material pile grid data, truck cargo hold grid data, basic static and dynamic information of the steel grabber, basic truck information, obstacle information, and profile information; The input data is stored in a scene structure; the scene structure contains all the elements of the scene and integrates this information to form a complete scene that can be used.

3. The intelligent decision-making task scheduling method for scrap steel yard grabbers according to claim 1 is characterized by: Use the following steps to determine if the truck is full: Input scene structure, which contains the truck cargo area and related scene information; Traverse the grid data of the warehouse, check the information in each grid, and determine whether it exceeds the height limit of the truck warehouse. If it exceeds, it means that the truck warehouse is full, otherwise it is not full.

4. The intelligent decision-making task scheduling method for a scrap steel yard steel grabber according to claim 1 is characterized in that: Step S2 of updating the grid data of the stockpile and truck cargo hold in the scene specifically includes the following steps: Input the grabbing information of the steel grabber after each grab, the information of the truck cargo hold, the mesh data of the stockpile, the mesh data of the truck cargo hold, and the profile information in the scene; Obtain updated mesh data and the results of steel grabbing by the grabber; Update the scene structure and apply the new data to the scene; Output the updated scene structure.

5. The intelligent decision-making task scheduling method for scrap steel yard grabbers according to claim 1 is characterized by: The placement point selection process is as follows: Input scene structure, which contains the information about the truck warehouse and material pile in the current scene; Determine whether this is the first time a placement point is selected. If so, select a placement point in front of the truck cargo hold, where the grid height does not exceed the height limit of the truck cargo hold. If this is not the first time, determine whether the height of the previous placement point has exceeded the height of the truck cargo hold. If not, continue to use the previous placement point. If the height has exceeded, select a new placement point to ensure that the height limit of the truck cargo hold is not exceeded. Output the placement point position selected in this round.

6. The intelligent decision-making task scheduling method for a scrap steel yard steel grabber according to claim 1 is characterized in that: The grab point selection process is as follows: Input scene structure; Determine whether the current grabbing window can be closed or ended. If so, calculate the grabbing position and determine the grabbing amount based on the material in the window. Calculate the time and path required to reach the grabbing position and determine whether the grabbing window is full. If the grabbing window is not full, calculate the grabbing percentage and related information. After determining whether all the capture windows are full or invalid, confirm whether it is necessary to switch the capture window, and finally output the result of the capture operation.

7. The intelligent decision-making task scheduling method for a steel grabber in a scrap steel yard according to claim 6 is characterized in that: The grab point selection is specific The following steps are involved: Step 1: Feasibility determination of the capture window: If feasible, go to step 2; otherwise, go to step 7. The feasibility determination of the window refers to: If the unit grid height variance is less than the preset threshold, that is, the surface of the entire pile is relatively flat, it is considered infeasible; If the height of the pile of molding materials from the ground is less than the preset threshold, making it difficult to grab, it is also considered unfeasible; Step 2: Calculate the grabbing position: First, determine the selection position: select the center of the rectangular area, calculate the ridge line at the center to determine the fine-grained orientation of the steel claw, and use the center of the ridge line as the grabbing position. The grabbing depth is based on the highest unit grid height. At the same time, the opening size is determined according to the shearing machine capacity limit. Step 3: Calculate the estimated weight of the profile within the window: Calculate the estimated profile volume as the weight based on the average height of the unit grid within the grasping range, the grasping depth and the size of the claw opening; Step 4: Calculate the Euclidean distance and time required for the claw to move to the grasping point; Step 5: Calculate the cost-effectiveness of capturing in this window. The calculation formula is as follows: R=W / T Where T represents the time required for the steel claw to move to the grasping position, W represents the estimated weight of the steel, and R represents the cost-effectiveness, that is, the greater the weight of steel grasped per unit time, the better; Step 6: Determine whether all windows have been calculated; if so, go to step 8; otherwise, go to step 7; Step 7: Window switching: Slide the window at a fixed step size and go to step 1; Step 8. Select the best cost-effective window: Based on the cost-effectiveness, select the best window corresponding to the steel grabbing position, depth, and steel claw direction.

8. An electronic device, characterized in that: including memory and processor; The memory is used to store computer programs; The processor is configured to implement the method according to any one of claims 1 to 7 when executing the computer program.

9. A computer-readable storage medium, wherein a computer program is stored on the storage medium, and when the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

10. A computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the computer program implements the method according to any one of claims 1 to 7.

Images