CNC machine tool feeding and discharging robot multi-mode motion control system and method

Abstract

The invention discloses a CNC machine tool feeding and discharging robot multi-mode motion control system and method, and relates to the technical field of robot motion control, and the method comprises the following steps: in the path planning process, firstly, according to the task requirement of each mechanical arm and the current working space state, the motion path of the mechanical arm is dynamically generated through real-time sensor feedback; after the complete motion path is generated, the path of the mechanical arm is divided into a plurality of sub-path sections according to a fixed time window, and each sub-path section represents a small continuous motion area of the mechanical arm. Through dynamic path prediction and conflict detection combined with real-time sensor feedback and a machine learning model, potential path conflicts can be recognized and evaluated in time, the time window and the task execution sequence can be intelligently adjusted according to conflict severity, it is ensured that each mechanical arm executes tasks within safe time and space, and the task execution efficiency is improved. The deadlock phenomenon is reduced, and the overall efficiency and stability are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of robot motion control, and particularly to a multi-modal motion control system and method for a CNC machine tool loading and unloading robot. Background Art

[0002] The multi-modal motion of a CNC machine tool loading and unloading robot refers to the ability of the robot to coordinate multiple motion modes during the automatic loading and unloading process of a CNC machine tool, and can switch or fuse different perception and motion methods according to task requirements to achieve efficient and precise operations. Specifically, multi-modal motion not only includes traditional mechanical motion modes such as linear, rotational, and interpolation, but also integrates data from various sensors such as vision, force sense, and displacement. Through intelligent algorithms, dynamic perception and adjustment of workpiece position, posture, clamping force, etc. are realized. For example, when grasping a workpiece, the robot can first roughly locate the position through vision, and then use a force sensor to precisely adjust the clamping force and angle to ensure stable clamping of the workpiece. This technology significantly improves the flexibility and intelligence level of loading and unloading, and is applicable to various complex processing environments.

[0003] In motion control, the path planning control system is mainly responsible for calculating and generating the optimal or feasible motion path from the starting point to the target point according to the robot task requirements and environmental constraints. This system combines the kinematic and dynamic models of the robot and sensor feedback information to ensure that the robot can accurately perform loading and unloading operations in a complex working environment. The role of the path planning control system is to optimize the robot's motion trajectory, avoid collisions with obstacles, and at the same time meet objectives such as time efficiency and energy consumption control, improving work precision and speed. Through intelligent path planning, the system can also adjust the motion path in real time to cope with different workpiece sizes, positions, and posture changes, ensuring that the robot can complete various tasks flexibly and efficiently during execution. Especially in multi-modal motion, it can effectively coordinate different motion methods and sensor inputs to ensure the safety and efficiency of operations.

[0004] The existing technology has the following deficiencies:

[0005] In the case of a large number of workpieces with a chaotic distribution, the existing path planning control system may not be able to effectively predict and avoid path conflicts between multiple robotic arms. Especially when multiple workpieces are on the same plane, multiple robotic arms may approach the same position simultaneously, resulting in a deadlock phenomenon and thus unable to continue subsequent operations. Such problems not only cause the operation to stagnate, but may also lead to the following serious consequences:

[0006] Robotic arm collision: Since the path planning system fails to adjust the path in time, it may cause collisions between multiple robotic arms, thereby damaging the equipment and increasing the maintenance cost;

[0007] Workpiece damage: Collisions may cause the displacement of the workpiece position or impacts during the clamping process, which may lead to workpiece damage or misalignment, seriously affecting the machining accuracy.

[0008] The above information disclosed in the background art section is only used to enhance the understanding of the background of the present disclosure. Therefore, it may include information that does not constitute the prior art known to those of ordinary skill in the art. Summary of the Invention

[0009] The object of the present invention is to provide a multi-modal motion control system and method for a CNC machine tool loading and unloading robot. Through the combination of real-time sensor feedback and a machine learning model for dynamic path prediction and conflict detection, it can not only timely identify and evaluate potential path conflicts, but also intelligently adjust the time window and task execution order according to the severity of the conflicts, ensuring that each robotic arm performs tasks within a safe time and space. The application of time delay technology effectively avoids multiple robotic arms entering the same area simultaneously, reduces the occurrence of deadlock phenomena, improves the overall efficiency and stability, and thus ensures the continuity of the production line and the machining accuracy of the workpiece, so as to solve the problems in the above background art.

[0010] To achieve the above object, the present invention provides the following technical solution: A multi-modal motion control method for a CNC machine tool loading and unloading robot, including the following steps:

[0011] During the path planning process, first, according to the task requirements of each robotic arm and the current working space state, the motion path of the robotic arm is dynamically generated through real-time sensor feedback;

[0012] After generating the complete motion path, the path of the robotic arm is divided into several sub-path segments according to a fixed time window, and each sub-path segment represents a small continuous motion area of the robotic arm;

[0013] When a robotic arm generates a new sub-path segment, a pre-trained machine learning model is used to predict conflicts for this sub-path segment;

[0014] When it is detected that there are path conflicts in the new sub-path segment, the severity of the conflicts is evaluated according to the prediction results of the machine learning model, and based on the evaluation results of the severity of the path conflicts, the original fixed time window is divided into several sub-time periods;

[0015] Within each sub-time period, time delay technology is used to control the execution order and timing of tasks, ensuring that only one robotic arm is allowed to execute the task corresponding to the motion path of each sub-time period.

[0016] Preferably, according to the task requirements of each robotic arm and the current working space state, the motion path of the robotic arm is dynamically generated through real-time sensor feedback, and the specific steps are as follows:

[0017] Continuously collect the surrounding environment data through the sensors integrated on each robotic arm;

[0018] Based on the real-time collected environment data, construct a dynamic environment model, perceive the current environment state through the environment model and generate a feasible preliminary path to ensure that the robotic arm can execute tasks in a dynamic environment;

[0019] After the preliminary path is generated, optimize the path according to the current workspace state to ensure that the path of the robotic arm is always in the optimal state during the execution process.

[0020] Preferably, when the robotic arm generates a new sub-path segment, collect the path data of all sub-path segments within this fixed time window. After preprocessing the collected path data, extract the features reflecting the conflict between this sub-path segment and other paths. Among them, the extracted features include the overlapping degree between the current sub-path segment and other paths and the time crossing degree between the current sub-path segment and other paths. Under the fixed time window, through feature engineering techniques, deeply analyze the extracted features, and respectively generate a path overlap index and a time crossing index, and quantify the conflict risk situation between the current sub-path segment and other paths through the path overlap index and the time crossing index.

[0021] Preferably, input the feature vector composed of the path overlap index and the time crossing index into the pre-trained machine learning model. Output the conflict coefficient through the machine learning model, and based on the conflict coefficient, conduct conflict prediction on this sub-path segment to determine whether the new sub-path segment has the risk of path conflict with other path segments.

[0022] Preferably, compare and analyze the conflict coefficient generated when using the pre-trained machine learning model to conduct conflict prediction on this sub-path segment with the pre-set conflict coefficient reference threshold to determine whether the new sub-path segment has the risk of path conflict with other path segments. The specific judgment steps are as follows:

[0023] If the conflict coefficient generated when conducting conflict prediction on this sub-path segment is greater than the conflict coefficient reference threshold, it is determined that the new sub-path segment has the risk of path conflict with other path segments; if the conflict coefficient generated when conducting conflict prediction on this sub-path segment is less than or equal to the conflict coefficient reference threshold, it is determined that the new sub-path segment has no risk of path conflict with other path segments.

[0024] Preferably, when it is detected that the new sub-path segment has a path conflict, divide the original fixed time window into several sub-time periods. The specific steps are as follows:

[0025] When a path conflict is detected in a newly detected sub-path segment, first evaluate the severity of the conflict according to the conflict coefficient predicted by the machine learning model. The conflict severity evaluation is completed through the following formula:

[0026]

[0027] , where: represents the conflict severity score, which is used to quantify the risk level of the path conflict, is the conflict coefficient predicted by the model, which reflects the conflict risk between the path segment and other paths, is the conflict coefficient reference threshold, which represents the set conflict risk tolerance;

[0028] According to the conflict severity evaluation result, that is, the conflict severity score , divide the original fixed time window into multiple sub-time periods, and each sub-time period corresponds to a task execution stage. The specific division process is as follows:

[0029]

[0030] , where: represents the length of the new sub-time period. The length of each sub-time period is dynamically adjusted according to the severity of the conflict, is the length of the original fixed time window.

[0031] Preferably, within each sub-time period, use time delay technology to control the execution order and timing of tasks to ensure that only one robotic arm is allowed to execute tasks for the motion path corresponding to each sub-time period. The specific steps are as follows:

[0032] Within each sub-time period, first, according to the conflict severity evaluation result, assign the task execution order and priority of each robotic arm, and dynamically adjust the task execution time of each robotic arm according to the length and task urgency of different sub-time periods to ensure that only one robotic arm executes tasks within each sub-time period;

[0033] Use time delay technology to precisely control the task execution order to ensure that the task execution times of each robotic arm do not overlap;

[0034] During the execution process, monitor the risk of path conflict in real time. If the execution path of a certain robotic arm has a potential conflict with the tasks of other robotic arms, adjust the parameters of the time delay to ensure that only one robotic arm is active within the conflict area.

[0035] Preferably, under the fixed time window, the specific steps for generating the path overlap index through feature engineering technology to deeply analyze the overlap degree between the current sub-path segment and other paths are as follows:

[0036] First, the overlapping area is initially calculated through the intersection of path shapes. Let the trajectory of each path segment be represented as a set of coordinate points. The overlapping degree between the current sub-path segment and other paths is calculated by the following formula:

[0037]

[0038] , where: represents the overlapping area between the current sub-path segment and other paths, represents the path and 's spatial intersection, and are the discretized representations of the path segments, is used to adjust the weight of the path within the specified area (e.g., the length, complexity, or speed of the path), and these factors will affect the risk of path overlap;

[0039] According to the calculation result of the overlapping area, a path overlap index is generated through the spatial occupancy and overlapping area of the path to quantify the conflict risk between the current sub-path segment and other paths. The generation expression of the path overlap index is:

[0040]

[0041] , where: represents the path overlap index, is the occupied area of the path within the workspace, represents the total area occupied by all paths .

[0042] Preferably, under a fixed time window, the specific steps to generate a time intersection index through in-depth analysis of the time intersection degree between the current sub-path segment and other paths by feature engineering techniques are as follows:

[0043] First, analyze the time intersection degree between the current sub-path segment and other paths within a fixed time window through real-time path data. For this purpose, define the time intersection interval as the intersection of the time intervals of two path segments. Let the time interval of the first path segment be and the time interval of the second path segment be , where , are the start and end times of the current sub-path segment, , are the start and end times of other path segments. If there is an intersection between these two time intervals, the length of the time intersection interval is expressed as:

[0044]

[0045] , in the formula: Represents the time intersection interval of two path segments, which measures the degree of temporal overlap between the two path segments. If the time length of the time intersection interval is long, it indicates a higher temporal overlap between the two path segments and a greater risk of conflict;

[0046] After identifying the time intersection interval, calculate the time intersection index through the length of the time intersection interval to quantify the conflict situation between the current sub-path segment and other paths. Suppose within a fixed time window, the length of the time intersection interval of path i is Then the lengths of the time intersection intervals of the current sub-path segment with all other paths can be respectively denoted as where n represents the total number of the current sub-path segment and other path segments, that is, how many other paths the current sub-path segment has a time intersection with within the fixed time window. The expression for calculating the time intersection index through the time intersection interval is:

[0047]

[0048] where: is the time intersection index, is the weighting factor, reflecting the priority of path i or the severity of the conflict (for example, a path with a higher priority may have a greater weight), T is the total length of the fixed time window, is the length of the time intersection interval of path i at time t.

[0049] The multi-modal motion control system for the CNC machine tool loading and unloading robot includes a path generation module, a path division module, a conflict prediction module, a time window division module, and a time delay module;

[0050] The path generation module dynamically generates the motion path of the robotic arm according to the task requirements of each robotic arm and the current working space state through real-time sensor feedback;

[0051] The path division module divides the path of the robotic arm into several sub-path segments according to a fixed time window, and each sub-path segment represents a small continuous motion area of the robotic arm;

[0052] The conflict prediction module, when the robotic arm generates a new sub-path segment, uses a pre-trained machine learning model to predict the conflict of the sub-path segment;

[0053] The time window division module, when detecting that there is a path conflict in the new sub-path segment, evaluates the severity of the conflict according to the prediction result of the machine learning model, and based on the evaluation result of the severity of the path conflict, divides the original fixed time window into several sub-time periods;

[0054] Time delay module, within each sub - time period, uses time delay technology to control the execution order and timing of tasks, ensuring that only one robotic arm is allowed to execute tasks for the motion path corresponding to each sub - time period.

[0055] In the above - mentioned technical solution, the technical effects and advantages provided by the present invention are as follows:

[0056] Through the combination of real - time sensor feedback and machine learning models for dynamic path prediction and conflict detection, the present invention can not only timely identify and evaluate potential path conflicts, but also intelligently adjust the time window and task execution order according to the severity of the conflicts, ensuring that each robotic arm executes tasks within a safe time and space. The application of time delay technology effectively avoids multiple robotic arms entering the same area simultaneously, reduces the occurrence of deadlock phenomena, improves the overall efficiency and stability, and thus ensures the continuity of the production line and the machining accuracy of workpieces. Brief Description of the Drawings

[0057] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the following will briefly introduce the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments recorded in the present invention. For those of ordinary skill in the art, other drawings can also be obtained based on these drawings.

[0058] Figure 1 It is a method flow chart of the multi - modal motion control method for the loading and unloading robot of the CNC machine tool of the present invention.

[0059] Figure 2 It is a module schematic diagram of the multi - modal motion control system for the loading and unloading robot of the CNC machine tool of the present invention. Detailed Embodiments

[0060] Now, the exemplary embodiments will be described more comprehensively with reference to the drawings. However, the exemplary embodiments can be implemented in various forms and should not be construed as limited to the examples set forth herein; rather, these exemplary embodiments are provided so that the present disclosure will be more comprehensive and complete, and the concept of the exemplary embodiments will be fully conveyed to those skilled in the art.

[0061] The present invention provides a multi - modal motion control method for the loading and unloading robot of the CNC machine tool as Figure 1 shown, including the following steps:

[0062] During the path planning process, first, according to the task requirements of each robotic arm and the current working space state, the motion path of the robotic arm is dynamically generated through real - time sensor feedback;

[0063] In the first step of path planning, each robotic arm first performs real-time perception of the surrounding environment through integrated sensors (such as vision sensors, depth cameras, force sensors, etc.). These sensors continuously collect environmental data, such as the position of workpieces, the dynamic changes of obstacles, the current position and state of the robotic arm, etc. After data collection, the sensors transmit the information to the central control system, which uses this real-time data to construct the current environmental model. The key to this process is to ensure the real-time and accuracy of the data, providing a basis for subsequent path planning. Through precise real-time data collection, the system can dynamically understand the changes in the workspace, ensuring that the path planning can adapt to environmental changes in a timely manner.

[0064] Based on the real-time collected environmental data, the system will construct a dynamic environmental model. This model usually includes the current position of the robotic arm, obstacles in the workspace, workpiece positions, and other elements that may affect path planning. Then, the path planning system will use this information to generate a preliminary motion path. Considering the requirements of the task (such as grasping, transporting, or placing), the robotic arm must avoid colliding with obstacles and try to select the optimal path as much as possible. The preliminary path planning will combine kinematic constraints and task requirements to determine a reasonable motion trajectory for the robotic arm. Environmental modeling ensures that the system can accurately perceive the current state and generate a feasible preliminary path, ensuring that the robotic arm can perform tasks in a dynamic environment.

[0065] After the preliminary path is generated, the system will optimize the path according to the current state of the workspace. At this time, the system will consider multiple factors, such as the shortest path, obstacle avoidance efficiency, smoothness of the robotic arm's movement, etc. To cope with sudden environmental changes (such as the movement of other robotic arms or changes in workpiece positions), the path planning will be adjusted in real time. This adjustment is achieved through a sensor feedback mechanism and dynamic optimization algorithms (such as A algorithm, D algorithm, etc.), ensuring that the path of the robotic arm is always in an optimal state during execution. Real-time path optimization and adjustment can handle uncertain environmental changes, ensuring that the robotic arm can perform tasks flexibly and efficiently and avoid possible path conflicts or collisions.

[0066] After generating the complete motion path, the path of the robotic arm is divided into several sub-path segments according to a fixed time window, and each sub-path segment represents a small continuous motion area of the robotic arm;

[0067] A sub-path segment is usually a straight or curved segment in a path. There are mainly two bases for dividing sub-path segments: one is to consider the physical characteristics of the path (such as displacement, speed, and acceleration limits); the other is according to the complexity of the task (such as grasping, handling, placing, etc.). In each sub-path segment, the movement of the robotic arm is relatively stable, facilitating subsequent collision detection and path optimization. By dividing the path into sub-path segments, the system can perform path analysis and optimization segment by segment, thereby improving the accuracy and flexibility of path planning.

[0068] Considering the physical characteristics of the path (such as displacement, speed, and acceleration limits) and the complexity of the task (such as grasping, handling, placing, etc.) is to ensure that the path division can adapt to the movement ability of the robotic arm and the task requirements. The physical characteristics determine the movement limits of the robotic arm, ensuring that the divided sub-path segments do not exceed the movement ability range of the robotic arm, thus avoiding collisions, jitters, or mistakes caused by overly fast or large movements. The complexity of the task determines the way of path division. Complex tasks such as grasping and placing usually require more refined path planning to ensure that the robotic arm can accurately reach the target position and complete the operation. Therefore, considering these two factors helps to ensure that the path division not only conforms to the movement ability of the robotic arm but also meets the high-precision requirements of task execution, thereby improving the overall operation efficiency and accuracy.

[0069] When the robotic arm generates a new sub-path segment, the path planning system will use a pre-trained machine learning model to predict conflicts for this sub-path segment;

[0070] When the robotic arm generates a new sub-path segment, collect the path data of all sub-path segments in this fixed time window. After preprocessing the collected path data, extract the features that reflect the conflict between this sub-path segment and other paths. Among them, the extracted features include the overlap degree between the current sub-path segment and other paths and the time crossing degree between the current sub-path segment and other paths. In the fixed time window, after deeply analyzing the extracted features through feature engineering techniques, generate a path overlap index and a time crossing index respectively, and quantify the conflict risk situation between the current sub-path segment and other paths through the path overlap index and the time crossing index;

[0071] A higher degree of overlap between the current sub-path segment and other paths usually indicates a greater risk of path conflict between the current sub-path segment and other paths. The reason is that when multiple paths overlap in space, the robotic arm may occupy the same area when executing these paths, resulting in collisions or interferences. Path overlap increases the likelihood of the robotic arm entering the same space at the same time, thus significantly increasing the risk of path conflict. On the contrary, if the degree of path overlap is low, it means that the movement trajectories and space occupations of the robotic arms are more dispersed, and the possibility of conflict is small. Therefore, the degree of path overlap directly reflects the potential spatial competition between robotic arms. The more overlap, the higher the risk of conflict.

[0072] The specific steps for generating the path overlap index through feature engineering techniques for the degree of overlap between the current sub-path segment and other paths under a fixed time window are as follows:

[0073] First, the overlapping area is initially calculated through the intersection of path shapes. Let the trajectory of each path segment be represented as a set of coordinate points. The degree of overlap between the current sub-path segment and other paths is calculated by the following formula:

[0074]

[0075] , where: represents the overlapping area between the current sub-path segment and other paths, represents the path and 's spatial intersection, and are the discretized representations of the path segments, is used to adjust the weight of the path in the specified area (for example, the length, complexity, or speed of the path), and these factors will affect the risk of path overlap;

[0076] Through the calculation of path intersection and the evaluation of space occupation, the initial degree of path overlap is obtained, which reflects the areas where conflicts may occur between paths.

[0077] According to the calculation result of the overlapping area, a path overlap index is generated through the space occupation and overlapping area of the path to quantify the conflict risk between the current sub-path segment and other paths. The generation expression of the path overlap index is:

[0078]

[0079] , where: represents the path overlap index, is the occupied area of the path in the workspace, represents all paths 's total occupied area;

[0080] This formula quantifies the risk of path overlap by calculating the ratio of the overlapping area to the total occupied area of all paths. Specifically, when the path occupied area is large and the overlapping regions are numerous, the overlap index is high, indicating a greater risk of conflict between paths. Conversely, if there is less path overlap, the conflict risk is smaller. Through the optimized path overlap index, the system can clearly reflect the degree of spatial conflict between paths.

[0081] As can be seen from the path overlap index, under a fixed time window, the larger the performance value of the path overlap index generated by deeply analyzing the overlap degree between the current sub-path segment and other paths through feature engineering techniques, the greater the risk of path conflict between the current sub-path segment and other paths. The reason is that the path overlap index measures the risk of path conflict by calculating the spatial overlap area between paths. When the overlapping area between the current sub-path segment and other paths is large, it means that multiple paths are active in the same area simultaneously, increasing the probability of their collision or interference. Therefore, when the path overlap index value is high, it indicates a greater conflict risk between these paths. When the overlap index value is small, it means that the degree of overlap between paths is low and the probability of conflict is small. Therefore, the level of the path overlap index directly reflects the severity of the potential conflict between paths.

[0082] When the time crossing degree between the current sub-path segment and other paths is high, it indicates a greater risk of path conflict between the current sub-path segment and other paths. The reason is that when multiple paths cross within the same time period, the robotic arm may enter the same working area simultaneously, resulting in spatial overlap and interference, thus increasing the risk of collision. If the paths do not overlap or overlap less in time, the operation times of the robotic arm in the same area are staggered, thereby reducing the possibility of path conflict. In short, a high time crossing degree means that the execution times of the paths are close or overlapping, increasing the possibility of path conflict; conversely, it means a lower chance of conflict and the paths can be executed safely.

[0083] Under a fixed time window, the specific steps for generating the time crossing index by deeply analyzing the time crossing degree between the current sub-path segment and other paths through feature engineering techniques are as follows:

[0084] First, analyze the time crossing degree between the current sub-path segment and other paths within a fixed time window through real-time path data. For this purpose, define the time crossing interval as the intersection of the time intervals of the two path segments. Let the time interval of the first path segment be and the time interval of the second path segment be where and are the start and end times of the current sub-path segment, and are the start and end times of other path segments. If there is an intersection between these two time intervals, the length of the time intersection interval is expressed as:

[0085]

[0086] , where: represents the time intersection interval of two path segments, which measures the degree of overlap of the two path segments in time. If the time length of the time intersection interval is long, it indicates a higher overlap in time between the two path segments and a greater risk of conflict.

[0087] The purpose of this step is to quantify the degree of intersection of two path segments in the time dimension and provide basic data for subsequent conflict assessment.

[0088] After identifying the time intersection interval, calculate the time intersection index through the length of the time intersection interval to quantify the conflict situation between the current sub-path segment and other paths. Suppose that within a fixed time window, the length of the time intersection interval of path i is , then the lengths of the time intersection intervals of the current sub-path segment with all other paths can be respectively denoted as , where n represents the total number of the current sub-path segment and other path segments, that is, how many other paths the current sub-path segment has time intersection with within the fixed time window. The expression for calculating the time intersection index through the time intersection interval is:

[0089]

[0090] , where: is the time intersection index, is the weighting factor, which reflects the priority of path i or the severity of the conflict (for example, a path with a higher priority may have a greater weight), T is the total length of the fixed time window, is the length of the time intersection interval of path i at time t;

[0091] Through integral calculation, the time intersection index provides a quantified conflict risk assessment value, quantifies the overall conflict risk through integral, and provides a basis for path optimization and conflict avoidance.

[0092] As can be seen from the time intersection index, under a fixed time window, the larger the performance value of the time intersection index generated by deeply analyzing the time intersection degree between the current sub-path segment and other paths through feature engineering technology, the greater the risk of path conflict between the current sub-path segment and other path segments. The reason is that the time intersection index reflects the intersection area in time by quantifying the time overlap degree between the current sub-path segment and other path segments within a fixed time window. If the intersection time is long, it means that multiple path segments occupy the same space or working area within the same time period, thus increasing the risk of collision or interference. Therefore, when the time intersection index is high, it means that the time overlap is more serious and the possibility of path conflict is also greater; on the contrary, a lower time intersection index indicates less time overlap between paths and a smaller conflict risk.

[0093] When the robotic arm generates a new sub-path segment, the feature vector composed of the path overlap index and the time intersection index is input into the pre-trained machine learning model. The machine learning model outputs a conflict coefficient, and based on the conflict coefficient, conflict prediction is performed on this sub-path segment to determine whether there is a risk of path conflict between the new sub-path segment and other path segments.

[0094] The pre-trained machine learning model refers to that in the path planning system, the model is trained through historical data or simulated data to optimize its learning process, enabling it to accurately predict path conflicts in actual work. In the context of this problem, the main task of this machine learning model is to predict whether there is a risk of conflict between different path segments based on the input feature vector (such as the path overlap index and the time intersection index), specifically reflected in outputting a conflict coefficient. The training process is based on known path data, which includes the time, space features of different path segments and their conflict situations. The model continuously adjusts its internal weight parameters through these data, thereby learning to identify the relationships between which path features and conflict risks.

[0095] Machine learning models are usually trained through a series of supervised learning or unsupervised learning algorithms (such as regression analysis, decision trees, neural networks, support vector machines, etc.). The training data includes different path overlap situations and time intersection situations, as well as the actually observed path conflict results (whether a collision or interference has occurred). Through these data, the model gradually adjusts its parameters to minimize the error between the prediction result and the actual situation, thus enabling the ability to predict new path segments. The training process of this model not only depends on the input static features (such as the geometric shape and time points of the path), but also considers dynamically changing factors, such as the speed and acceleration of the robotic arm. As the training progresses, the model can gradually understand which factors are the key to causing path conflicts under different working conditions. Therefore, the trained model can more efficiently and accurately predict new path conflicts in actual operations.

[0096] In the path planning system, the role of the pre-trained machine learning model is mainly reflected in the analysis and processing of the input feature vector. When the robotic arm generates a new sub-path segment, the system calculates the path overlap index and the time intersection index based on the spatial overlap degree and the time intersection situation between the current path and other paths, synthesizes these feature values into a feature vector, and then inputs this vector into the trained machine learning model. Through the output of the model, that is, the conflict coefficient, the system can determine whether there is a risk of conflict between this sub-path segment and other path segments.

[0097] In practical applications, the conflict coefficient, as the output of the machine learning model, represents the probability or risk degree of path conflict. The higher the value of this conflict coefficient, the greater the conflict risk between the current path segment and other path segments. On the contrary, if the conflict coefficient is low, it can be considered that the execution risk of this path segment is small. Through continuous learning and optimization, the model can quickly predict path conflicts based on similar situations in historical data when facing new path data and give corresponding conflict evaluation values. Through the application of this machine learning model, the path planning system can not only handle complex path conflict problems in multi-robotic arm cooperation tasks but also cope with uncertain and dynamically changing working environments. As the system continuously accumulates new path data, the prediction ability of the machine learning model will also continuously improve, further enhancing the path conflict early warning ability.

[0098] The machine learning model is not specifically limited here, and any machine learning model that can realize the comprehensive analysis of the path overlap index and the time intersection index to generate the conflict coefficient is acceptable. To implement the technical solution of the present invention, the present invention provides a specific implementation manner; the expression for generating the conflict coefficient is: , where , are the weight coefficients of the path overlap index and the time intersection index respectively, and , are both greater than 0. The weight coefficient refers to the parameter used to adjust the influence degree of different features in the model. Specifically, and are the weight coefficients of the path overlap index ( ) and the time intersection index ( ) respectively. They are used to quantify the influence of these features on the generated conflict coefficient ( Degree of contribution. By adjusting these weight coefficients, the model can learn which features (such as path overlap and time intersection) are more important in predicting path conflicts. The magnitudes of the weight coefficients directly affect the calculation result of the final conflict coefficient, thus affecting the prediction accuracy of path conflicts. During the model training process, the weight coefficients are optimized through learning algorithms to ensure that the model can generate accurate conflict predictions based on the input features.

[0099] From the conflict coefficient, it can be seen that under a fixed time window, the larger the performance value of the path overlap index generated by deeply analyzing the overlap degree between the current sub-path segment and other paths through feature engineering techniques, and the larger the performance value of the time intersection index generated by deeply analyzing the time intersection degree between the current sub-path segment and other paths through feature engineering techniques, that is, the larger the performance value of the conflict coefficient generated when using the pre-trained machine learning model to perform conflict prediction on this sub-path segment, the greater the risk of path conflict between the current sub-path segment and other paths. Conversely, it indicates that the risk of path conflict between the current sub-path segment and other paths is smaller.

[0100] Compare and analyze the conflict coefficient generated when using the pre-trained machine learning model to perform conflict prediction on this sub-path segment with the pre-set conflict coefficient reference threshold to determine whether there is a risk of path conflict between the new sub-path segment and other path segments. The specific judgment steps are as follows:

[0101] If the conflict coefficient generated when performing conflict prediction on this sub-path segment is greater than the conflict coefficient reference threshold, it is determined that there is a risk of path conflict between the new sub-path segment and other path segments; if the conflict coefficient generated when performing conflict prediction on this sub-path segment is less than or equal to the conflict coefficient reference threshold, it is determined that there is no risk of path conflict between the new sub-path segment and other path segments.

[0102] When it is detected that there is a path conflict in the new sub-path segment, evaluate the severity of the conflict according to the prediction result of the machine learning model, and based on the evaluation result of the severity of the path conflict, divide the original fixed time window into several sub-time periods;

[0103] When it is detected that there is a path conflict in the new sub-path segment, divide the original fixed time window into several sub-time periods. The specific steps are as follows:

[0104] When it is detected that there is a path conflict in the new sub-path segment, first evaluate the severity of the conflict according to the conflict coefficient predicted by the machine learning model. The conflict severity evaluation is completed through the following formula:

[0105]

[0106] , where: Indicates the conflict severity score, which is used to quantify the risk level of path conflicts. Is the conflict coefficient predicted by the model, reflecting the conflict risk between the path segment and other paths. Is the reference threshold of the conflict coefficient, representing the set conflict risk tolerance.

[0107] Through this step, the conflict coefficient can be standardized with the reference threshold of the conflict coefficient to obtain a ratio, indicating the severity of the conflict. When the conflict severity score has a large value, it indicates that the path conflict is relatively serious, and the system will take more strict path adjustment and time window division for this sub-path segment.

[0108] According to the conflict severity assessment result, that is, the conflict severity score , the original fixed time window is divided into multiple sub-time periods, and each sub-time period corresponds to a task execution stage. The specific division process is as follows:

[0109]

[0110] , where: Represents the length of the new sub-time period. The length of each sub-time period is dynamically adjusted according to the severity of the conflict. Is the length of the original fixed time window.

[0111] By dynamically adjusting the time window through the conflict severity score, the system can precisely control the execution time of the robotic arm according to the complexity of the task and the risk of path conflicts. More serious conflicts result in the time window being divided into multiple sub-time periods, which helps to avoid multiple robotic arms entering the same area to execute tasks in the same time period, thus effectively reducing the occurrence of path conflicts.

[0112] Within each sub-time period, time delay technology is used to control the execution order and timing of tasks, ensuring that only one robotic arm is allowed to execute tasks on the motion path corresponding to each sub-time period.

[0113] The specific steps are as follows:

[0114] Step 1: Task scheduling and time window allocation;

[0115] Within each sub-time period, first, according to the conflict severity assessment result, assign the task execution order and priority of each robotic arm. Dynamically adjust the task execution time of each robotic arm according to the length of different sub-time periods and the urgency of the tasks, ensuring that only one robotic arm executes tasks within each sub-time period.

[0116] Step 2: Introduction of time delay and execution synchronization;

[0117] To avoid multiple robotic arms performing tasks simultaneously in the same area, a time-delay technique is adopted to precisely control the task execution sequence. The time delay is dynamically calculated based on the current state and path requirements of each robotic arm, ensuring that the task execution times of each robotic arm do not overlap.

[0118] Step 3: Path conflict monitoring and adjustment;

[0119] During the execution process, the risk of path conflicts is monitored in real time. If the execution path of a certain robotic arm potentially conflicts with the tasks of other robotic arms, by adjusting the parameters of the time delay, it is ensured that only one robotic arm is active in the conflict area, thus avoiding the cross-execution of tasks.

[0120] Through the above solution, the path planning accuracy and conflict avoidance ability in multi-robotic arm collaboration are effectively improved, thus significantly reducing the risk of robotic arm collisions and workpiece damage. Through the combination of real-time sensor feedback and machine learning models for dynamic path prediction and conflict detection, not only can potential path conflicts be identified and evaluated in a timely manner, but also the time window and task execution sequence can be intelligently adjusted according to the severity of the conflict, ensuring that each robotic arm performs tasks within a safe time and space. The application of the time-delay technique effectively avoids multiple robotic arms entering the same area simultaneously, reduces the occurrence of deadlock phenomena, improves the efficiency and stability of the overall system, and thus ensures the continuity of the production line and the machining accuracy of workpieces.

[0121] The method for capturing the flight state of a flying object provided by the embodiment of the present invention is implemented through the above-mentioned system for capturing the flight state of a flying object. The specific methods and processes of the system for capturing the flight state of a flying object are detailed in the embodiments of the method for capturing the flight state of a flying object above, and will not be elaborated here.

[0122] The above formulas are all dimensionless and take their numerical calculations. The formulas are obtained by collecting a large amount of data for software simulation to obtain a formula closest to the actual situation. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0123] Only some exemplary embodiments of the present invention have been described by way of illustration above. Undoubtedly, for those of ordinary skill in the art, without departing from the spirit and scope of the present invention, the described embodiments can be modified in various different ways. Therefore, the above drawings and descriptions are illustrative in nature and should not be construed as limiting the scope of protection of the claims of the present invention.

[0124] As described above, it is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should all be covered within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claimed rights.

Claims

1. A multi-modal motion control method for a CNC machine tool loading and unloading robot, characterized in that: The following steps are involved: During the path planning process, the motion path of each robot arm is dynamically generated through real-time sensor feedback according to the task requirements of each robot arm and the current workspace status; After the complete motion path is generated, the path of the robot is divided into several sub-path segments according to a fixed time window, and each sub-path segment represents a small continuous motion area of ​​the robot; When the robot generates a new sub-path segment, the pre-trained machine learning model is used to predict the conflict of the sub-path segment; When a new sub-path segment is detected to have a path conflict, the severity of the conflict is evaluated according to the prediction results of the machine learning model, and based on the severity evaluation results of the path conflict, the original fixed time window is divided into several sub-time periods; In each sub-time period, time delay technology is used to control the execution order and timing of tasks, ensuring that the motion path corresponding to each sub-time period only allows one robot arm to perform the task.

2. The multi-modal motion control method of a CNC machine tool loading and unloading robot according to claim 1, characterized in that: According to the task requirements of each robot and the current workspace status, the motion path of the robot is dynamically generated through real-time sensor feedback. The specific steps are as follows: The surrounding environment data is continuously collected through the sensors integrated on each robotic arm; Based on the real-time collected environmental data, a dynamic environmental model is constructed. The environmental model is used to perceive the current environmental state and generate a feasible preliminary path to ensure that the robot arm can perform tasks in a dynamic environment. After the preliminary path is generated, the path is optimized according to the current workspace status to ensure that the path of the robot is always in the optimal state during execution.

3. The multi-modal motion control method of a CNC machine tool loading and unloading robot according to claim 1, characterized in that: When the robot generates a new sub-path segment, the path data of all sub-path segments in the fixed time window is collected. After preprocessing the collected path data, features reflecting the conflict between the sub-path segment and other paths are extracted. The extracted features include the degree of overlap between the current sub-path segment and other paths and the degree of time intersection between the current sub-path segment and other paths. In the fixed time window, after in-depth analysis of the extracted features through feature engineering technology, path overlap indicators and time intersection indicators are generated respectively. The path overlap indicators and time intersection indicators are used to quantify the conflict risk between the current sub-path segment and other paths.

4. The multi-modal motion control method of a CNC machine tool loading and unloading robot according to claim 3 is characterized in that: The feature vector composed of path overlap index and time intersection index is input into the pre-trained machine learning model. The conflict coefficient is output by the machine learning model. The conflict prediction is performed on the sub-path segment based on the conflict coefficient to determine whether the new sub-path segment has the risk of path conflict with other path segments.

5. The multi-modal motion control method of a CNC machine tool loading and unloading robot according to claim 4, characterized in that: The conflict coefficient generated when the pre-trained machine learning model is used to predict the conflict of the sub-path segment is compared and analyzed with the pre-set reference threshold of the conflict coefficient to determine whether the new sub-path segment has the risk of path conflict with other path segments. The specific judgment steps are as follows: If the conflict coefficient generated when predicting the conflict for the sub-path segment is greater than the conflict coefficient reference threshold, it is determined that the new sub-path segment has a risk of path conflict with other path segments; if the conflict coefficient generated when predicting the conflict for the sub-path segment is less than or equal to the conflict coefficient reference threshold, it is determined that the new sub-path segment has no risk of path conflict with other path segments.

6. The multi-modal motion control method of a CNC machine tool loading and unloading robot according to claim 5, characterized in that: When a new sub-path segment is detected to have a path conflict, the original fixed time window is divided into several sub-time periods. The specific steps are as follows: When a new sub-path segment is detected to have a path conflict, the severity of the conflict is evaluated based on the conflict coefficient predicted by the machine learning model. The conflict severity evaluation is completed using the following formula: in: Represents the conflict severity score, which is used to quantify the risk of path conflict. is the conflict coefficient predicted by the model, reflecting the conflict risk between the path segment and other paths. is the reference threshold of the conflict coefficient; Based on the conflict severity assessment results, i.e., the conflict severity score , the original fixed time window is divided into multiple sub-time periods, each of which corresponds to a task execution phase. The specific division process is as follows: in: Indicates the length of the new sub-time period. The length of each sub-time period is dynamically adjusted according to the severity of the conflict. is the original fixed time window length.

7. The multi-modal motion control method of a CNC machine tool loading and unloading robot according to claim 1, characterized in that: In each sub-time period, the time delay technology is used to control the execution order and timing of the tasks to ensure that the motion path corresponding to each sub-time period only allows one robot arm to perform the task. The specific steps are as follows: In each sub-time period, firstly, the task execution order and priority of each robot arm are assigned according to the conflict severity assessment results. According to the length of different sub-time periods and the urgency of the tasks, the task execution time of each robot arm is dynamically adjusted to ensure that only one robot arm executes the task in each sub-time period. The time delay technology is used to precisely control the order of task execution to ensure that the task execution time of each robot arm does not overlap; During the execution process, the risk of path conflict is monitored in real time. If the execution path of a certain robot arm has a potential conflict with the tasks of other robot arms, the time delay parameters are adjusted to ensure that only one robot arm is active in the conflict area.

8. The multi-modal motion control method of a CNC machine tool loading and unloading robot according to claim 3, characterized in that: In a fixed time window, the specific steps of generating path overlap indicators by deeply analyzing the overlap between the current sub-path segment and other paths through feature engineering technology are as follows: First, a preliminary calculation of the overlapping area is performed through the intersection of the path shapes; According to the calculation results of the overlapping area, the path overlap index is generated by the spatial occupancy and overlapping area of ​​the path to quantify the conflict risk between the current sub-path segment and other paths.

9. The multi-modal motion control method of a CNC machine tool loading and unloading robot according to claim 3, characterized in that: Under a fixed time window, the specific steps for generating a time intersection index by deeply analyzing the time intersection degree of the current sub-path segment and other paths through feature engineering technology are as follows: Analyze the time intersection degree between the current sub-path segment and other paths within a fixed time window through real-time path data; After identifying the time intersection interval, the time intersection index is calculated by the length of the time intersection interval to quantify the conflict between the current sub-path segment and other paths.

10. A multi-modal motion control system for a CNC machine tool loading and unloading robot, used to implement the multi-modal motion control method for a CNC machine tool loading and unloading robot as described in any one of claims 1 to 9, characterized in that: It includes a path generation module, a path division module, a conflict prediction module, a time window division module and a time delay module; The path generation module dynamically generates the motion path of each robot arm through real-time sensor feedback according to the task requirements of each robot arm and the current workspace status; The path division module divides the path of the robot into several sub-path segments according to a fixed time window. Each sub-path segment represents a small continuous motion area of ​​the robot. The conflict prediction module uses the pre-trained machine learning model to predict conflicts for a new sub-path segment when the robot generates one; The time window division module, when a new sub-path segment has a path conflict, evaluates the severity of the conflict according to the prediction results of the machine learning model, and divides the original fixed time window into several sub-time segments based on the severity evaluation results of the path conflict; The time delay module uses time delay technology to control the execution order and timing of tasks in each sub-time period, ensuring that the motion path corresponding to each sub-time period only allows one robot arm to perform the task.

Images