CN120080323B - CNC machine tool feeding and discharging robot multi-mode motion control system and method - Google Patents

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

Info

Publication number
CN120080323B
CN120080323B CN202510561148.0A CN202510561148A CN120080323B CN 120080323 B CN120080323 B CN 120080323B CN 202510561148 A CN202510561148 A CN 202510561148A CN 120080323 B CN120080323 B CN 120080323B
Authority
CN
China
Prior art keywords
path
sub
time
conflict
mechanical arm
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN202510561148.0A
Other languages
Chinese (zh)
Other versions
CN120080323A (en
Inventor
吕军
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Dongguan Juwei Electronic Technology Co ltd
Original Assignee
Dongguan Juwei Electronic Technology Co ltd
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Dongguan Juwei Electronic Technology Co ltd filed Critical Dongguan Juwei Electronic Technology Co ltd
Priority to CN202510561148.0A priority Critical patent/CN120080323B/en
Publication of CN120080323A publication Critical patent/CN120080323A/en
Application granted granted Critical
Publication of CN120080323B publication Critical patent/CN120080323B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
    • B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
    • B25J9/00—Programme-controlled manipulators
    • B25J9/16—Programme controls
    • B25J9/1656—Programme controls characterised by programming, planning systems for manipulators
    • B25J9/1664—Programme controls characterised by programming, planning systems for manipulators characterised by motion, path, trajectory planning
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B23—MACHINE TOOLS; METAL-WORKING NOT OTHERWISE PROVIDED FOR
    • B23Q—DETAILS, COMPONENTS, OR ACCESSORIES FOR MACHINE TOOLS, e.g. ARRANGEMENTS FOR COPYING OR CONTROLLING; MACHINE TOOLS IN GENERAL CHARACTERISED BY THE CONSTRUCTION OF PARTICULAR DETAILS OR COMPONENTS; COMBINATIONS OR ASSOCIATIONS OF METAL-WORKING MACHINES, NOT DIRECTED TO A PARTICULAR RESULT
    • B23Q7/00—Arrangements for handling work specially combined with or arranged in, or specially adapted for use in connection with, machine tools, e.g. for conveying, loading, positioning, discharging, sorting
    • B23Q7/04—Arrangements for handling work specially combined with or arranged in, or specially adapted for use in connection with, machine tools, e.g. for conveying, loading, positioning, discharging, sorting by means of grippers
    • B—PERFORMING OPERATIONS; TRANSPORTING
    • B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
    • B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
    • B25J9/00—Programme-controlled manipulators
    • B25J9/16—Programme controls
    • B25J9/1628—Programme controls characterised by the control loop
    • B25J9/163—Programme controls characterised by the control loop learning, adaptive, model based, rule based expert control
    • G—PHYSICS
    • G06—COMPUTING OR CALCULATING; COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning

Landscapes

  • Engineering & Computer Science (AREA)
  • Mechanical Engineering (AREA)
  • Software Systems (AREA)
  • Robotics (AREA)
  • Theoretical Computer Science (AREA)
  • Evolutionary Computation (AREA)
  • Medical Informatics (AREA)
  • Physics & Mathematics (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Data Mining & Analysis (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Artificial Intelligence (AREA)
  • Numerical Control (AREA)
  • Feedback Control In General (AREA)

Abstract

The invention discloses a multi-mode motion control system and method for a feeding and discharging robot of a CNC machine tool, which relate to the technical field of robot motion control and comprise the following steps that in a path planning process, a motion path of a mechanical arm is dynamically generated through real-time sensor feedback according to task requirements and current working space states of each mechanical arm; after the complete motion path is generated, the path of the mechanical arm is divided into a plurality of sub-path segments according to a fixed time window, and each sub-path segment represents a small continuous motion area of the mechanical arm. According to the invention, through the dynamic path prediction and conflict detection combined with the real-time sensor feedback and the machine learning model, potential path conflicts can be timely identified and evaluated, a time window and a task execution sequence can be intelligently adjusted according to conflict severity, each mechanical arm is ensured to execute tasks in safe time and space, the occurrence of deadlock phenomenon is reduced, and the overall efficiency and stability are improved.

Description

CNC machine tool feeding and discharging robot multi-mode motion control system and method
Technical Field
The invention relates to the technical field of robot motion control, in particular to a multi-mode motion control system and method for a feeding and discharging robot of a CNC machine tool.
Background
The multi-mode motion of the feeding and discharging robot of the CNC machine tool means that the robot has coordination capability of multiple motion modes in the automatic feeding and discharging process of the CNC machine tool, and different perception and motion modes can be switched or fused according to task requirements so as to realize efficient and accurate operation. Specifically, the multi-mode motion not only comprises traditional mechanical motion modes such as straight line, rotation, interpolation and the like, but also fuses data from various sensors such as vision, force sense, displacement and the like, and realizes dynamic sensing and adjustment of workpiece positions, postures, clamping forces and the like through an intelligent algorithm. For example, when grabbing the work piece, the robot can be used for accurately adjusting the clamping force and angle through visual positioning of the approximate position, and then the force sensor is used for guaranteeing stable clamping of the work piece. The technology remarkably improves the flexibility and the intelligence level of feeding and discharging, and is suitable for various complex processing environments.
In motion control, the path planning control system is mainly responsible for calculating and generating an optimal or feasible motion path from a starting point to a target point according to the task requirements and environmental constraints of the robot. The system combines the kinematics and dynamics model of the robot and the feedback information of the sensor, so that the robot can accurately execute feeding and discharging operations in a complex working environment. The path planning control system has the functions of optimizing the motion trail of the robot, avoiding collision with obstacles, meeting the targets of time efficiency, energy consumption control and the like, and improving the working precision and speed. Through intelligent path planning, the system can also adjust the motion path in real time, cope with the changes of the sizes, the positions and the postures of different workpieces, ensure that the robot flexibly and efficiently completes various tasks in the execution process, and particularly in multi-mode motion, can effectively coordinate different motion modes and sensor input, and ensure the safety and the high efficiency of operation.
The prior art has the following defects:
Under the condition that the number of the workpieces is numerous and the distribution is disordered, the existing path planning control system can not effectively predict and avoid path conflicts among the plurality of mechanical arms, and particularly when the plurality of workpieces are located on the same plane, the plurality of mechanical arms can approach the same position at the same time, so that a deadlock phenomenon is caused, and subsequent operation can not be continued. Such problems not only lead to job stalls, but can also have serious consequences:
the collision of the mechanical arms, namely that the path planning system can not adjust the path in time, so that the mutual collision among a plurality of mechanical arms can be caused, thereby damaging equipment and increasing maintenance cost;
the workpiece is damaged, namely the collision can cause the deviation of the position of the workpiece or the impact in the clamping process, so that the workpiece is damaged or misplaced, and the machining precision is seriously affected.
The above information disclosed in the background section is only for enhancement of understanding of the background of the disclosure and therefore it may include information that does not form the prior art that is already known to a person of ordinary skill in the art.
Disclosure of Invention
The invention aims to provide a multi-mode motion control system and method for a CNC machine tool feeding and discharging robot, which can not only timely identify and evaluate potential path conflicts, but also intelligently adjust a time window and a task execution sequence according to conflict severity by means of dynamic path prediction and conflict detection combined with a machine learning model through real-time sensor feedback, and ensure that each mechanical arm executes tasks in safe time and space. The application of the time delay technology effectively avoids a plurality of mechanical arms from entering the same area at the same time, reduces the occurrence of deadlock phenomenon, and improves the overall efficiency and stability, thereby ensuring the continuity of the production line and the processing precision of workpieces, and solving the problems in the background technology.
In order to achieve the purpose, the invention provides the following technical scheme that the multi-mode motion control method for the feeding and discharging robot of the CNC machine tool comprises the following steps:
In the path planning process, firstly, dynamically generating a motion path of each mechanical arm through real-time sensor feedback according to the task requirement of each mechanical arm and the current working space state;
After a complete motion path is generated, dividing the path of the mechanical arm into a plurality of sub-path sections according to a fixed time window, wherein each sub-path section represents a small continuous motion area of the mechanical arm;
When the mechanical arm generates a new sub-path segment, carrying out conflict prediction on the sub-path segment by utilizing a machine learning model which is trained in advance;
when detecting that a new sub-path segment has path conflict, evaluating the severity of the conflict according to a prediction result of a machine learning model, and dividing an original fixed time window into a plurality of sub-time segments based on the severity evaluation result of the path conflict;
in each sub-time period, the execution sequence and the time of the tasks are controlled by adopting a time delay technology, so that the motion path corresponding to each sub-time period only allows one mechanical arm to execute the tasks.
Preferably, according to the task requirement and the current working space state of each mechanical arm, the motion path of the mechanical arm is dynamically generated through real-time sensor feedback, and the specific steps are as follows:
Continuously acquiring surrounding environment data through sensors integrated on each mechanical arm;
based on the environmental data collected in real time, a dynamic environmental model is constructed, the current environmental state is perceived through the environmental model, a feasible preliminary path is generated, and the mechanical arm is ensured to execute tasks in the dynamic environment;
After the preliminary path is generated, path optimization is performed according to the current working space state, so that the path of the mechanical arm is always in an optimal state in the execution process.
Preferably, when the mechanical arm generates a new sub-path segment, path data of all sub-path segments under the fixed time window are collected, after the collected path data are preprocessed, features reflecting conflict between the sub-path segment and other paths are extracted, wherein the extracted features comprise the overlapping degree of the current sub-path segment and other paths and the time crossing degree of the current sub-path segment and other paths, under the fixed time window, after the extracted features are subjected to deep analysis through a feature engineering technology, path overlapping indexes and time crossing indexes are respectively generated, and the conflict risk situation of the current sub-path segment and other paths is quantified through the path overlapping indexes and the time crossing indexes.
Preferably, feature vectors composed of path overlapping indexes and time crossing indexes are input into a pre-trained machine learning model, a conflict coefficient is output through the machine learning model, conflict prediction is carried out on the sub-path segments based on the conflict coefficient, and whether the new sub-path segments have path conflict risks with other path segments or not is judged.
Preferably, the collision coefficient generated when the pre-trained machine learning model is used for carrying out collision prediction on the sub-path segment is compared with a preset collision coefficient reference threshold value for analysis, and whether the new sub-path segment has the risk of path collision with other path segments is judged, wherein the specific judgment steps are as follows:
If the conflict coefficient generated when the conflict prediction is carried out on the sub-path segment is larger than the conflict coefficient reference threshold, judging that the new sub-path segment and other path segments have the risk of path conflict, and if the conflict coefficient generated when the conflict prediction is carried out on the sub-path segment is smaller than or equal to the conflict coefficient reference threshold, judging that the new sub-path segment and other path segments have no risk of path conflict.
Preferably, when detecting that a path conflict exists in a new sub-path segment, dividing an original fixed time window into a plurality of sub-time segments, and specifically, the steps are as follows:
When a path conflict exists in a new sub-path segment, firstly, evaluating the severity of the conflict according to a conflict coefficient predicted by a machine learning model, wherein the evaluation of the severity of the conflict is completed by the following formula:
Wherein: Represents a collision severity score, for quantifying the risk level of path collisions, The collision coefficients predicted by the model reflect the collision risk of the path segment and other paths,A threshold is referenced for the conflict coefficient, and the set conflict risk tolerance is represented;
based on the result of the collision severity assessment, i.e. the collision severity score Dividing an original fixed time window into a plurality of sub-time periods, wherein each sub-time period corresponds to a task execution stage, and the specific dividing process is as follows:
Wherein: Representing the length of the new sub-period, the length of each sub-period is dynamically adjusted according to the severity of the collision, Is the original fixed time window length.
Preferably, in each sub-time period, the execution sequence and the time of the tasks are controlled by adopting a time delay technology, so that the motion path corresponding to each sub-time period only allows one mechanical arm to execute the tasks, and the specific steps are as follows:
In each sub-time period, firstly, according to a conflict severity evaluation result, distributing the task execution sequence and priority of each mechanical arm, and according to the lengths of different sub-time periods and task urgency, dynamically adjusting the task execution time of each mechanical arm, so as to ensure that only one mechanical arm executes the task in each sub-time period;
the task execution sequence is precisely controlled by adopting a time delay technology, so that the task execution time of each mechanical arm is ensured not to overlap;
In the execution process, the risk of path conflict is monitored in real time, and if the execution path of one mechanical arm has potential conflict with the tasks of other mechanical arms, only one mechanical arm in the conflict area is ensured to be active by adjusting the time delay parameter.
Preferably, under a fixed time window, the specific steps of generating the path overlapping index after the overlapping degree of the current sub-path segment and other paths is subjected to depth analysis by the feature engineering technology are as follows:
Firstly, carrying out preliminary calculation of an overlapping area through a path shape intersection, setting the track of each path segment to be a series of coordinate point sets, and calculating the overlapping degree of the current sub-path segment and other paths through the following formula:
Wherein: Representing the overlapping area of the current sub-path segment with other paths, Representing a pathAndIs defined by the spatial intersection of (a) and (b),AndIs a discretized representation of the path segment,For adjusting the weights of paths within a specified area (e.g., the length, complexity, or speed of the paths), these factors can affect the risk of path overlap;
generating a path overlapping index according to the calculation result of the overlapping area and through the space occupation and the overlapping area of the path, quantifying the conflict risk of the current sub-path segment and other paths, wherein the generation expression of the path overlapping index is as follows:
Wherein: The path overlap index is indicated as such, Is a pathThe occupied area in the working space is that,Representing all pathsThe total area occupied.
Preferably, under a fixed time window, the specific steps of generating the time crossing index after performing depth analysis on the time crossing degree of the current sub-path segment and other paths by using the feature engineering technology are as follows:
firstly, analyzing the time intersection degree of the current sub-path segment and other paths in a fixed time window through real-time path data, defining a time intersection interval as the intersection of time intervals of two path segments, and setting the time interval of the first path segment as And the time interval of the second path segment isWherein、Is the start and end time of the current sub-path segment,、Is the start and end time of the other path segment, if there is an intersection of these two time intervals, the length of the time crossing interval is expressed as:
Wherein: representing the time crossing interval of two path segments, the degree of overlap of the two path segments in time is measured. If the time length of the time crossing interval is longer, the time between the two path segments is higher in overlap, and the collision risk is also higher;
After the time crossing section is identified, calculating a time crossing index according to the time crossing section length, quantifying the conflict situation of the current sub-path section and other paths, and setting the time crossing section length of the path i as follows in a fixed time window The time-crossing interval length of the current sub-path segment and all other paths can be respectively recorded asWherein 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 crossing in a fixed time window, and the expression of calculating the time crossing index through the time crossing interval is:
Wherein: as an indicator of the time-crossing, Reflecting the priority of path i or the severity of the collision (e.g., a higher priority path may be weighted more), T is the total length of the fixed time window,Is the time crossing interval length of the path i at time t.
The CNC machine tool feeding and discharging robot multi-mode motion control system comprises 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 a motion path of the mechanical arm through real-time sensor feedback according to the task requirement and the current working space state of each mechanical arm;
The path dividing module divides the path of the mechanical arm 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;
the conflict prediction module is used for carrying out conflict prediction on the sub-path segment by utilizing a machine learning model which is trained in advance when the mechanical arm generates a new sub-path segment;
The time window dividing module is used for estimating the severity of the collision according to the prediction result of the machine learning model when detecting that the new sub-path segment has the path collision, and dividing the original fixed time window into a plurality of sub-time segments based on the severity estimation result of the path collision;
And the time delay module is used for controlling the execution sequence and the time of the tasks by adopting a time delay technology in each sub-time period, so that the motion path corresponding to each sub-time period only allows one mechanical arm to execute the tasks.
In the technical scheme, the invention has the technical effects and advantages that:
According to the invention, through the dynamic path prediction and conflict detection combined with the machine learning model by the real-time sensor feedback, potential path conflicts can be timely identified and evaluated, and the time window and the task execution sequence can be intelligently adjusted according to the conflict severity, so that each mechanical arm can execute tasks in safe time and space. The application of the time delay technology effectively avoids a plurality of mechanical arms from entering the same area at the same time, reduces the occurrence of deadlock phenomenon, and improves the overall efficiency and stability, thereby ensuring the continuity of the production line and the processing precision of workpieces.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for the embodiments will be briefly described below, and it is apparent that the drawings in the following description are only some embodiments described in the present application, and other drawings may be obtained according to these drawings for those skilled in the art.
FIG. 1 is a flow chart of a method for controlling the multi-mode motion of a feeding and discharging robot of a CNC machine tool.
Fig. 2 is a schematic block diagram of a multi-mode motion control system of a feeding and discharging robot of a CNC machine.
Detailed Description
Example embodiments will now be described more fully with reference to the accompanying drawings. However, the example embodiments may be embodied in many different forms and should not be construed as limited to the examples set forth herein, but rather, the example embodiments are provided so that this disclosure will be more thorough and complete, and will fully convey the concept of the example embodiments to those skilled in the art.
The invention provides a multi-mode motion control method of a CNC machine tool feeding and discharging robot shown in fig. 1, which comprises the following steps:
In the path planning process, firstly, dynamically generating a motion path of each mechanical arm through real-time sensor feedback according to the task requirement of each mechanical arm and the current working space state;
In the first step of path planning, each mechanical arm first senses the surrounding environment in real time through integrated sensors (such as vision sensors, depth cameras, force sensors, etc.). The sensors can continuously collect surrounding environmental data such as workpiece positions, dynamic changes of obstacles, current positions and states of the mechanical arms and the like. After data acquisition, the sensors will transmit information to a central control system, which uses the real-time data to construct a current environmental model. The key of the process is to ensure the real-time performance and accuracy of the data and provide a basis for subsequent path planning. Through accurate real-time data acquisition, the system can dynamically know the change of the working space, and ensures that the path planning can adapt to the change of the environment in time.
Based on the environmental data collected in real time, the system will construct a dynamic environmental model. This model typically includes the current position of the robotic arm, obstructions in the workspace, workpiece positions, and other elements that may affect path planning. The path planning system then uses this information to generate a preliminary path of motion, and the robotic arm must avoid collision obstacles and choose the optimal path as much as possible, taking into account the task requirements (e.g., grasping, handling, or placement). The preliminary path planning combines the kinematic constraint and the task requirement to determine a reasonable motion trail for the mechanical arm. The environment modeling ensures that the system can accurately sense the current state and generate a feasible preliminary path, and ensures that the mechanical arm can execute tasks in a dynamic environment.
After the preliminary path is generated, the system performs path optimization according to the current working space state. At this time, the system may consider a plurality of factors, such as a shortest path, obstacle avoidance efficiency, smoothness of movement of the robot arm, and the like. To cope with sudden environmental changes (e.g., other robot movements or workpiece position changes), the path plan is adjusted in real time. The adjustment is realized through a sensor feedback mechanism and a dynamic optimization algorithm (such as an A algorithm, a D algorithm and the like), so that the path of the mechanical arm is always in an optimal state in the execution process. Real-time path optimization and adjustment can cope with uncertain environmental changes, ensure that the mechanical arm can flexibly and efficiently execute tasks, and avoid possible path collision or collision.
After a complete motion path is generated, dividing the path of the mechanical arm into a plurality of sub-path sections according to a fixed time window, wherein each sub-path section represents a small continuous motion area of the mechanical arm;
a sub-path segment is typically a straight line or curve in the path. The basis of dividing the sub-path segment is mainly two, namely considering physical characteristics (such as displacement, speed and acceleration limitation) of the path and according to the complexity (such as grabbing, carrying, placing and the like) of the task. In each sub-path segment, the motion 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 analyze and optimize the path segment by segment, thereby improving the accuracy and flexibility of path planning.
Physical characteristics of the path (such as displacement, speed and acceleration limitations) and task complexity (such as grasping, handling, placement, etc.) are considered to ensure that the path splitting can accommodate the motion capabilities and task requirements of the robotic arm. The physical characteristics determine the motion limit of the mechanical arm, and ensure that the divided sub-path segments do not exceed the motion capability range of the mechanical arm, so that collision, shaking or errors caused by too fast or too large motion are avoided. The complexity of the task, which determines the way the path is divided, typically requires finer path planning, such as grasping and placement, to ensure that the robotic arm can accurately reach the target location and complete the operation. Therefore, considering the two factors helps to ensure that the path division meets the motion capability of the mechanical arm and the high-precision requirement of task execution, thereby improving the overall working efficiency and precision.
When the mechanical arm generates a new sub-path segment, the path planning system carries out conflict prediction on the sub-path segment by utilizing a machine learning model which is trained in advance;
When the mechanical arm generates a new sub-path segment, collecting path data of all sub-path segments under the fixed time window, preprocessing the collected path data, extracting features reflecting conflict between the sub-path segment and other paths from the preprocessed path data, wherein the extracted features comprise the overlapping degree of the current sub-path segment and other paths and the time crossing degree of the current sub-path segment and other paths, and under the fixed time window, after the extracted features are subjected to deep analysis through a feature engineering technology, respectively generating a path overlapping index and a time crossing index, and quantifying the conflict risk condition of the current sub-path segment and other paths through the path overlapping index and the time crossing index;
A higher degree of overlap of the current sub-path segment with other paths generally indicates a greater risk of path collisions of the current sub-path segment with other paths. The reason is that when a plurality of paths are spatially overlapped, the robot arm may occupy the same area when performing the paths, resulting in collision or interference. The overlapping of paths increases the likelihood that the robotic arms enter the same space at the same time, thereby significantly increasing the risk of path collisions. Conversely, if the path overlapping degree is low, this means that the motion track and the space occupation of the mechanical arm are scattered, and the possibility of collision is low. Thus, the degree of overlap of the paths directly reflects the potential spatial competition between the robotic arms, with the more overlap, the higher the risk of collision.
Under a fixed time window, the specific steps of generating a path overlapping index after carrying out depth analysis on the overlapping degree of the current sub-path segment and other paths by using a characteristic engineering technology are as follows:
Firstly, carrying out preliminary calculation of an overlapping area through a path shape intersection, setting the track of each path segment to be a series of coordinate point sets, and calculating the overlapping degree of the current sub-path segment and other paths through the following formula:
Wherein: Representing the overlapping area of the current sub-path segment with other paths, Representing a pathAndIs defined by the spatial intersection of (a) and (b),AndIs a discretized representation of the path segment,For adjusting the weights of paths within a specified area (e.g., the length, complexity, or speed of the paths), these factors can affect the risk of path overlap;
The preliminary path overlapping degree is obtained through calculation of the path intersection and evaluation of space occupation, and the possible conflict area between paths is reflected.
Generating a path overlapping index according to the calculation result of the overlapping area and through the space occupation and the overlapping area of the path, quantifying the conflict risk of the current sub-path segment and other paths, wherein the generation expression of the path overlapping index is as follows:
Wherein: The path overlap index is indicated as such, Is a pathThe occupied area in the working space is that,Representing all pathsThe total area occupied;
This formula quantifies the risk of paths overlapping by calculating the ratio of the overlap area to the total footprint of all paths. In particular, when the path occupation area is large and the overlapping area is large, the overlapping index is high, meaning that the collision risk between paths is large. Otherwise, if the paths overlap less, the collision risk is smaller. The system can clearly reflect the space conflict degree between the paths through the optimized path overlapping index.
The method is characterized in that the larger the performance value of the path overlapping index generated after the overlapping degree of the current sub-path segment and other paths is subjected to the depth analysis by the feature engineering technology under the fixed time window, the larger the risk of the path collision between the current sub-path segment and the other paths is indicated, and the path overlapping index is used for measuring the risk of the path collision by calculating the space overlapping area between the paths. When the overlapping area of the front sub-path segment and other paths is large, it means that multiple paths are active simultaneously in the same area, and the probability of collision or interference of the paths is increased. Therefore, when the path overlap index value is high, it indicates that there is a large risk of collision between these paths. And when the overlap index value is smaller, the overlap degree between paths is lower, and the probability of collision is smaller. Thus, the high or low level of the path overlap indicator directly reflects the severity of potential conflicts between paths.
When the time crossing degree of the current sub-path segment and other paths is higher, the greater the risk that the current sub-path segment has path conflict with other paths is indicated. The reason is that when multiple paths intersect within the same time period, the robotic arms may enter the same working area at the same time, resulting in spatial overlap and interference, thereby increasing the risk of collision. If the paths do not overlap in time or overlap less, the operating times of the robotic arms in the same region are staggered, thereby reducing the likelihood of path collisions. In short, when the time crossing degree is high, the execution time of the paths approaches or overlaps, so that the possibility of path collision is increased, otherwise, the possibility of collision is lower, and the paths can be safely executed.
Under a fixed time window, the specific steps of generating a time crossing index after carrying out depth analysis on the time crossing degree of the current sub-path segment and other paths by using a characteristic engineering technology are as follows:
firstly, analyzing the time intersection degree of the current sub-path segment and other paths in a fixed time window through real-time path data, defining a time intersection interval as the intersection of time intervals of two path segments, and setting the time interval of the first path segment as And the time interval of the second path segment isWherein、Is the start and end time of the current sub-path segment,、Is the start and end time of the other path segment, if there is an intersection of these two time intervals, the length of the time crossing interval is expressed as:
Wherein: Representing the time crossing interval of two path segments, the degree of overlap of the two path segments in time is measured. If the time length of the time crossing interval is longer, this means that there is a higher overlap in time between the two path segments and the risk of collision is also greater.
The purpose of this step is to quantify the degree of intersection of two path segments in the time dimension, providing the underlying data for subsequent collision assessment.
After the time crossing section is identified, calculating a time crossing index according to the time crossing section length, quantifying the conflict situation of the current sub-path section and other paths, and setting the time crossing section length of the path i as follows in a fixed time windowThe time-crossing interval length of the current sub-path segment and all other paths can be respectively recorded asWherein 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 crossing in a fixed time window, and the expression of calculating the time crossing index through the time crossing interval is:
Wherein: as an indicator of the time-crossing, Reflecting the priority of path i or the severity of the collision (e.g., a higher priority path may be weighted more), T is the total length of the fixed time window,The time crossing interval length of the path i at the time t;
the time cross index provides a quantized conflict risk assessment value through integral calculation, and the integral conflict risk is quantized through integral, so that a basis is provided for path optimization and conflict avoidance.
According to the time crossing index, under a fixed time window, the larger the expression value of the time crossing index generated after the time crossing degree of the current sub-path segment and other paths is subjected to deep analysis by the feature engineering technology, the larger the risk of the path conflict between the current sub-path segment and other path segments is indicated. The reason is that the time crossing index represents the intersection region of the current sub-path segment with other path segments in time by quantifying the degree of time overlap of them within a fixed time window. If the intersection time is longer, it means that multiple path segments occupy the same space or working area within the same time period, thereby increasing the risk of collision or interference. Therefore, when the time crossing index is higher, the time overlapping is more serious, and the probability of path collision is higher, and conversely, when the time crossing index is lower, the time overlapping among paths is less, and the collision risk is smaller.
When the mechanical arm generates a new sub-path segment, the feature vector composed of the path overlapping index and the time crossing index is input into a pre-trained machine learning model, a conflict coefficient is output through the machine learning model, the sub-path segment is subjected to conflict prediction based on the conflict coefficient, and whether the new sub-path segment has a path conflict risk with other path segments or not is judged.
The machine learning model which is trained in advance is that in a path planning system, the model is trained through historical data or simulation data, and the learning process is optimized, so that the path conflict can be accurately predicted in actual work. In the context of the present problem, the main task of this machine learning model is to predict whether there is a risk of collision between different path segments according to the input feature vectors (such as the path overlap index and the time intersection index), which is embodied in outputting a collision coefficient. The training process is based on known path data, which includes temporal, spatial features of the different path segments and their collision conditions, through which the model constantly adjusts its internal weight parameters, thereby learning which path features are related to the collision risk.
Machine learning models are typically trained by a series of supervised or unsupervised learning algorithms (e.g., regression analysis, decision trees, neural networks, support vector machines, etc.). The training data includes different path overlap conditions and time crossing conditions, as well as the actual observed path collision results (whether collisions or disturbances occur). Through the data, the model gradually adjusts the parameters of the model to minimize the error between the prediction result and the real situation, so that the model has the capability of predicting a new path segment. The training process of the model not only depends on the input static characteristics (such as the geometry of the path and the time point), but also considers the factors of dynamic change, such as the speed, the acceleration and the like of the mechanical arm. With the deep training, the model can gradually understand which factors are key to causing the path conflict under different working conditions, so that the trained model can more efficiently and accurately predict the new path conflict in actual operation.
In a path planning system, the function of a pre-trained machine learning model is mainly embodied in the analysis and processing of input feature vectors. When the mechanical arm generates a new sub-path segment, the system calculates a path overlapping index and a time crossing index according to the space overlapping degree and the time crossing condition of the current path and other paths, synthesizes the characteristic values into a characteristic vector, and then inputs the vector into a machine learning model after training. Through the output of the model, i.e., the collision coefficient, the system can determine whether the sub-path segment collides with other path segments.
In practical applications, the collision coefficient is taken as an output of the machine learning model and represents the probability or risk degree of path collision. The higher the value of the collision coefficient, the greater the risk of collision of the current path segment with other path segments. Conversely, if the collision coefficient is low, the execution risk of the path segment is considered to be low. The model can predict path conflict according to similar conditions in the historical data rapidly when facing new path data through continuous learning and optimization, and gives out corresponding conflict evaluation values. By applying the machine learning model, the path planning system can not only handle complex path conflict problems in multi-mechanical arm cooperation tasks, but also can handle uncertain and dynamically changed working environments. Along with the continuous accumulation of new path data of the system, the prediction capability of the machine learning model is also continuously improved, and the path conflict early warning capability is further enhanced.
The machine learning model is not particularly limited herein, and can implement path overlap indexAnd time crossing indexComprehensive analysis is carried out to generate conflict coefficientThe invention provides a specific implementation mode for realizing the technical proposal of the invention, namely the machine learning model of the inventionThe generated expression is: In which, in the process, 、Respectively, path overlap indexAnd time crossing indexAnd (2) weight coefficient of、Are all greater than 0. The weight coefficients refer to parameters for adjusting the degree of influence of different features in the model. In particular the number of the elements,AndRespectively the path overlap index) And time crossing index [ ]) Weight coefficient of (c) in the above-mentioned formula (c). They are used to quantify the conflicting coefficients generated by these feature pairs) Is a contribution degree of (2). By adjusting these weighting coefficients, the model can learn which features (e.g., path overlap and time crossing) are more important in predicting path collisions. The magnitude of the weight coefficient directly influences the calculation result of the final collision coefficient, thereby influencing the prediction accuracy of the path collision. In the model training process, the weight coefficient is optimized through a learning algorithm, so that the model can generate accurate conflict prediction according to the input characteristics.
According to the conflict coefficient, under a fixed time window, the larger the performance value of the path overlapping index generated after the overlapping degree of the current sub-path section and other paths is subjected to the deep analysis by the feature engineering technology, the larger the performance value of the time crossing index generated after the time crossing degree of the current sub-path section and other paths is subjected to the deep analysis by the feature engineering technology, namely, the larger the performance value of the conflict coefficient generated when the sub-path section is subjected to the conflict prediction by utilizing the machine learning model which is trained in advance, the larger the risk of the path conflict between the current sub-path section and other paths is shown, and otherwise, the smaller the risk of the path conflict between the current sub-path section and other paths is shown.
Comparing and analyzing a conflict coefficient generated when the machine learning model which is trained in advance is used for carrying out conflict prediction on the sub-path segment with a preset conflict coefficient reference threshold value, and judging whether a new sub-path segment has a risk of path conflict with other path segments or not, wherein the specific judging steps are as follows:
If the conflict coefficient generated when the conflict prediction is carried out on the sub-path segment is larger than the conflict coefficient reference threshold, judging that the new sub-path segment and other path segments have the risk of path conflict, and if the conflict coefficient generated when the conflict prediction is carried out on the sub-path segment is smaller than or equal to the conflict coefficient reference threshold, judging that the new sub-path segment and other path segments have no risk of path conflict.
When detecting that a new sub-path segment has path conflict, evaluating the severity of the conflict according to a prediction result of a machine learning model, and dividing an original fixed time window into a plurality of sub-time segments based on the severity evaluation result of the path conflict;
when detecting that a new sub-path segment has path conflict, dividing an original fixed time window into a plurality of sub-time segments, wherein the specific steps are as follows:
When a path conflict exists in a new sub-path segment, firstly, evaluating the severity of the conflict according to a conflict coefficient predicted by a machine learning model, wherein the evaluation of the severity of the conflict is completed by the following formula:
Wherein: Represents a collision severity score, for quantifying the risk level of path collisions, The collision coefficients predicted by the model reflect the collision risk of the path segment and other paths,A threshold is referenced for the conflict coefficient, and the set conflict risk tolerance is represented;
Through this step, the conflict factor and the conflict factor reference threshold can be normalized to obtain a ratio, which indicates the severity of the conflict. When conflict severity scores When the value of (a) is larger, indicating that the path conflict is more serious, the system will take more strict path adjustment and time window division for the sub-path segment.
Based on the result of the collision severity assessment, i.e. the collision severity scoreDividing an original fixed time window into a plurality of sub-time periods, wherein each sub-time period corresponds to a task execution stage, and the specific dividing process is as follows:
Wherein: Representing the length of the new sub-period, the length of each sub-period is dynamically adjusted according to the severity of the collision, The length of the fixed time window is set as the original value;
By dynamically adjusting the time window through the conflict severity score, the system can accurately control the execution time of the mechanical arm according to the complexity of the task and the risk of the path conflict. The time window is subdivided into a plurality of sub-time periods due to serious conflicts, so that the plurality of mechanical arms can be prevented from entering the same area to execute tasks in the same time period, and the occurrence of path conflicts is effectively reduced.
In each sub-time period, the execution sequence and the time of the tasks are controlled by adopting a time delay technology, so that the motion path corresponding to each sub-time period only allows one mechanical arm to execute the tasks;
The specific steps are as follows:
Step 1, task scheduling and time window allocation;
and in each sub-time period, firstly, distributing the task execution sequence and the priority of each mechanical arm according to the conflict severity evaluation result. And dynamically adjusting the task execution time of each mechanical arm according to the lengths of different sub-time periods and the task urgency, so as to ensure that only one mechanical arm executes the task in each sub-time period.
Step 2, time delay introduction and execution synchronization;
In order to avoid that a plurality of mechanical arms execute tasks simultaneously in the same area, a time delay technology is adopted to accurately 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.
Step 3, path conflict monitoring and adjusting;
During execution, the risk of path collisions is monitored in real time. If the execution path of one mechanical arm has potential conflict with the tasks of other mechanical arms, only one mechanical arm in the conflict area is ensured to be active by adjusting the time delay parameter, so that the cross execution of the tasks is avoided.
By the aid of the scheme, path planning precision and collision avoidance capability in multi-mechanical arm cooperation are effectively improved, and therefore risks of mechanical arm collision and workpiece damage are remarkably reduced. The dynamic path prediction and conflict detection combined with the machine learning model through the real-time sensor feedback can not only timely identify and evaluate potential path conflicts, but also intelligently adjust a time window and a task execution sequence according to the conflict severity, so that each mechanical arm can execute tasks in safe time and space. The application of the time delay technology effectively avoids a plurality of mechanical arms from entering the same area at the same time, reduces the occurrence of deadlock, and improves the efficiency and stability of the whole system, thereby ensuring the continuity of the production line and the processing precision of workpieces.
The method for capturing the flying state of the flying object provided by the embodiment of the invention is realized through the system for capturing the flying state of the flying object, and the specific method and the flow of the system for capturing the flying state of the flying object are detailed in the embodiment of the method for capturing the flying state of the flying object, and are not repeated herein.
The above formulas are all formulas with dimensions removed and numerical values calculated, the formulas are formulas with a large amount of data collected for software simulation to obtain the latest real situation, and preset parameters in the formulas are set by those skilled in the art according to the actual situation.
While certain exemplary embodiments of the present invention have been described above by way of illustration only, it will be apparent to those of ordinary skill in the art that modifications may be made to the described embodiments in various different ways without departing from the spirit and scope of the invention. Accordingly, the drawings and description are to be regarded as illustrative in nature and not as restrictive of the scope of the invention, which is defined by the appended claims.
The foregoing is merely illustrative of the present application, and the present application is not limited thereto, and any person skilled in the art will readily recognize that variations or substitutions are within the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims (8)

  1. The multi-mode motion control method of the CNC machine tool feeding and discharging robot is characterized by comprising the following steps of:
    in the path planning process, dynamically generating a motion path of each mechanical arm through real-time sensor feedback according to the task requirement of each mechanical arm and the current working space state;
    After a complete motion path is generated, dividing the path of the mechanical arm into a plurality of sub-path sections according to a fixed time window, wherein each sub-path section represents a small continuous motion area of the mechanical arm;
    When the mechanical arm generates a new sub-path segment, carrying out conflict prediction on the sub-path segment by utilizing a machine learning model which is trained in advance;
    when detecting that a new sub-path segment has path conflict, evaluating the severity of the conflict according to a prediction result of a machine learning model, and dividing an original fixed time window into a plurality of sub-time segments based on the severity evaluation result of the path conflict;
    in each sub-time period, the execution sequence and the time of the tasks are controlled by adopting a time delay technology, so that the motion path corresponding to each sub-time period only allows one mechanical arm to execute the tasks;
    When the mechanical arm generates a new sub-path segment, collecting path data of all sub-path segments under the fixed time window, preprocessing the collected path data, extracting features reflecting conflict between the sub-path segment and other paths from the preprocessed path data, wherein the extracted features comprise the overlapping degree of the current sub-path segment and other paths and the time crossing degree of the current sub-path segment and other paths, and under the fixed time window, after the extracted features are subjected to deep analysis through a feature engineering technology, respectively generating a path overlapping index and a time crossing index, and quantifying the conflict risk condition of the current sub-path segment and other paths through the path overlapping index and the time crossing index;
    when detecting that a new sub-path segment has path conflict, dividing an original fixed time window into a plurality of sub-time segments, wherein the specific steps are as follows:
    when a path conflict exists in the new sub-path segment, evaluating the severity of the conflict according to a conflict coefficient predicted by a machine learning model, wherein the evaluation of the severity of the conflict is completed by the following formula: Wherein: Represents a collision severity score, for quantifying the risk level of path collisions, Reflecting the collision risk of the path segment and other paths for the collision coefficient predicted by the model,Referencing a threshold for a collision coefficient;
    based on the result of the collision severity assessment, i.e. the collision severity score Dividing an original fixed time window into a plurality of sub-time periods, wherein each sub-time period corresponds to a task execution stage, and the specific dividing process is as follows: Wherein: Representing the length of the new sub-period, the length of each sub-period is dynamically adjusted according to the severity of the collision, Is the original fixed time window length.
  2. 2. The method for multi-modal motion control of a CNC machine tool loading and unloading robot according to claim 1, wherein the motion path of the robot is dynamically generated by real-time sensor feedback according to the task requirement and the current working space state of each robot, and the specific steps are as follows:
    Continuously acquiring surrounding environment data through sensors integrated on each mechanical arm;
    based on the environmental data collected in real time, a dynamic environmental model is constructed, the current environmental state is perceived through the environmental model, a feasible preliminary path is generated, and the mechanical arm is ensured to execute tasks in the dynamic environment;
    After the preliminary path is generated, path optimization is performed according to the current working space state, so that the path of the mechanical arm is always in an optimal state in the execution process.
  3. 3. The method for multi-modal motion control of a CNC machine tool loading and unloading robot according to claim 1, wherein a feature vector composed of a path overlap index and a time cross index is input into a machine learning model which is trained in advance, a collision coefficient is output through the machine learning model, collision prediction is performed on the sub-path segment based on the collision coefficient, and whether a new sub-path segment has a risk of path collision with other path segments is judged.
  4. 4. The method for multi-modal motion control of a CNC machine tool loading and unloading robot according to claim 3, wherein the step of comparing a collision coefficient generated when the sub-path segment is collision predicted by using a machine learning model which is trained in advance with a preset collision coefficient reference threshold value to analyze, and determining whether a new sub-path segment has a risk of collision with other path segments is as follows:
    If the conflict coefficient generated when the conflict prediction is carried out on the sub-path segment is larger than the conflict coefficient reference threshold, judging that the new sub-path segment and other path segments have the risk of path conflict, and if the conflict coefficient generated when the conflict prediction is carried out on the sub-path segment is smaller than or equal to the conflict coefficient reference threshold, judging that the new sub-path segment and other path segments have no risk of path conflict.
  5. 5. The method for multi-mode motion control of a CNC machine feeding and discharging robot according to claim 1, wherein in each sub-time period, a time delay technique is adopted to control the execution sequence and the time of the tasks, so as to ensure that only one mechanical arm is allowed to execute the tasks by the motion path corresponding to each sub-time period, and the specific steps are as follows:
    In each sub-time period, firstly, according to a conflict severity evaluation result, distributing the task execution sequence and priority of each mechanical arm, and according to the lengths of different sub-time periods and task urgency, dynamically adjusting the task execution time of each mechanical arm, so as to ensure that only one mechanical arm executes the task in each sub-time period;
    the task execution sequence is precisely controlled by adopting a time delay technology, so that the task execution time of each mechanical arm is ensured not to overlap;
    In the execution process, the risk of path conflict is monitored in real time, and if the execution path of one mechanical arm has potential conflict with the tasks of other mechanical arms, only one mechanical arm in the conflict area is ensured to be active by adjusting the time delay parameter.
  6. 6. The method for controlling the multi-mode motion of the feeding and discharging robot of the CNC machine tool according to claim 1, wherein the specific steps of generating the path overlapping index after the overlapping degree of the current sub-path segment and other paths is deeply analyzed by the feature engineering technology under the fixed time window are as follows:
    Firstly, carrying out preliminary calculation of an overlapping area through a path shape intersection;
    and generating a path overlapping index according to the calculation result of the overlapping area and through the space occupation and the overlapping area of the path, and quantifying the conflict risk of the current sub-path segment and other paths.
  7. 7. The method for controlling the multi-mode motion of the feeding and discharging robot of the CNC machine tool according to claim 1, wherein the specific steps of generating the time intersection index after the depth analysis of the time intersection degree of the current sub-path segment and other paths by the feature engineering technology under the fixed time window are as follows:
    Analyzing the time intersection degree of the current sub-path segment and other paths in a fixed time window through real-time path data;
    After the time crossing interval is identified, calculating a time crossing index according to the length of the time crossing interval, and quantifying the conflict situation of the current sub-path segment and other paths.
  8. A CNC machine feeding and discharging robot multi-mode motion control system for implementing the CNC machine feeding and discharging robot multi-mode motion control method according to any one of the above claims 1 to 7, characterized by comprising a path generating module, a path dividing module, a collision predicting module, a time window dividing module and a time delay module;
    the path generation module dynamically generates a motion path of the mechanical arm through real-time sensor feedback according to the task requirement and the current working space state of each mechanical arm;
    The path dividing module divides the path of the mechanical arm 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;
    the conflict prediction module is used for carrying out conflict prediction on the sub-path segment by utilizing a machine learning model which is trained in advance when the mechanical arm generates a new sub-path segment;
    The time window dividing module is used for estimating the severity of the collision according to the prediction result of the machine learning model when detecting that the new sub-path segment has the path collision, and dividing the original fixed time window into a plurality of sub-time segments based on the severity estimation result of the path collision;
    And the time delay module is used for controlling the execution sequence and the time of the tasks by adopting a time delay technology in each sub-time period, so that the motion path corresponding to each sub-time period only allows one mechanical arm to execute the tasks.
CN202510561148.0A 2025-04-30 2025-04-30 CNC machine tool feeding and discharging robot multi-mode motion control system and method Active CN120080323B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202510561148.0A CN120080323B (en) 2025-04-30 2025-04-30 CNC machine tool feeding and discharging robot multi-mode motion control system and method

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202510561148.0A CN120080323B (en) 2025-04-30 2025-04-30 CNC machine tool feeding and discharging robot multi-mode motion control system and method

Publications (2)

Publication Number Publication Date
CN120080323A CN120080323A (en) 2025-06-03
CN120080323B true CN120080323B (en) 2025-08-05

Family

ID=95856412

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202510561148.0A Active CN120080323B (en) 2025-04-30 2025-04-30 CNC machine tool feeding and discharging robot multi-mode motion control system and method

Country Status (1)

Country Link
CN (1) CN120080323B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN120620233B (en) * 2025-08-13 2025-10-03 南京东新慧科信息技术有限公司 A quadruped robot path optimization method and system

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117301053A (en) * 2023-09-19 2023-12-29 西门子（中国）有限公司 Path control method, device and storage medium
CN118466505A (en) * 2024-05-21 2024-08-09 北京邮电大学 Multi-robot path planning method based on multi-objective optimization and DNN priority sorting

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117021086A (en) * 2023-08-11 2023-11-10 广域铭岛数字科技有限公司 Robot travel management method, system, electronic device and readable storage medium

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117301053A (en) * 2023-09-19 2023-12-29 西门子（中国）有限公司 Path control method, device and storage medium
CN118466505A (en) * 2024-05-21 2024-08-09 北京邮电大学 Multi-robot path planning method based on multi-objective optimization and DNN priority sorting

Also Published As

Publication number Publication date
CN120080323A (en) 2025-06-03

Similar Documents

Publication Publication Date Title
Zhang et al. Recurrent neural network for motion trajectory prediction in human-robot collaborative assembly
US10717196B2 (en) Machine learning device, robot system, and machine learning method for learning workpiece picking operation
CN102744727B (en) Method and apparatus for predicting interference between target section of robot and peripheral object
JP6652549B2 (en) Chip removal device and information processing device
US20170032283A1 (en) Machine learning unit, spindle replacement judging device, controller, machine tool, production system, and machine learning method, which are able to judge necessity of spindle replacement
CN120080323B (en) CNC machine tool feeding and discharging robot multi-mode motion control system and method
CN117519215B (en) Multi-AGV driving control method, device, equipment and storage medium
KR20200054327A (en) 3D space monitoring device, 3D space monitoring method, and 3D space monitoring program
WO2023173000A1 (en) Motion planning and control for robots in shared workspace employing staging poses
CN115146798A (en) Method and system for whole-process monitoring and assistance of assembly robot based on ontology data
CN117798934A (en) Multi-step autonomous assembly operation decision-making method of cooperative robot
CN118456126A (en) Polishing path optimization method of cladding layer on inner surface of cylindrical barrel based on machine learning
US10459424B2 (en) Numerical controller for controlling tapping
CN118605135A (en) A robot posture control method for nut feeding kit
CN118605538B (en) Multi-source-based target data collaborative processing method and system
CN120196046A (en) Real-time control method, system and device for machining path of axis-linked curved surface components
CN119472452A (en) Intelligent control system of bridge deck maintenance robot
CN118357931B (en) Path planning method and device for double-arm robot, electronic equipment and storage medium
CN119974015A (en) Method and device for automatically deploying working parameters of industrial robots and industrial robots
CN120116227A (en) Autonomous decision-making and control algorithm for industrial robots based on reinforcement learning
CN116330344A (en) A Collision Detection Method for Collaborative Robots Based on Supervised Learning Support Vector Machine
US20250249586A1 (en) Motion planning and control for robots in shared workspace employing look ahead planning
CN119604824A (en) Robust motion planning and/or control for multi-robot environments
CN120480895A (en) Industrial robot system based on high-precision control and intelligent operation and maintenance
Guo et al. Time-bound optimal planning in CNC machine considering machining safety

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant