Grinding machine preparation whole-process monitoring system based on computer vision

Abstract

The invention relates to the field of intelligent manufacturing, in particular to a grinding machine manufacturing full-process monitoring system based on computer vision, and aims at various angle grinder assembly components to control a carrying mechanical arm to carry out carrying verification detection based on a target path by obtaining the target path of a carrying task needing to be executed by the carrying mechanical arm. According to the method, a stress interference section and stress interference characteristics are determined, the carrying stress type of a single-type angle grinder assembly component is further determined, and then the carrying mechanical arm is monitored and controlled on the basis of the carrying stress type for the carrying stress type of the angle grinder assembly component. According to the angle grinder assembling assembly, due to the influence of inertia and internal shaking on the carrying process and the monitoring precision, carrying parameters and detection strategies are adaptively adjusted, the operation efficiency of a carrying mechanical arm is guaranteed, the monitoring precision of the carrying mechanical arm is guaranteed, and the stability and reliability of the angle grinder accessory assembling and carrying process are improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent manufacturing, and in particular to a computer vision-based monitoring system for the entire process of grinding machine manufacturing. Background Technology

[0002] With the development of automation technology, industrial robots have been widely used in various fields. For example, in the field of angle grinder manufacturing, sorting and transfer of angle grinder assembly components are required. Currently, visual perception can be used to perceive the task objects that the handling robot needs to perform, monitor the task progress, and then intervene in the handling robot's handling behavior, so that the handling robot can adapt to the working environment and ensure the accuracy of the handling robot's movements.

[0003] For example, Chinese Patent Publication No. CN112657864A discloses an intelligent sorting and sanding center for furniture component processing, including a sander, a guide rack, a positioning detection mechanism, a translational conveying mechanism, and a sorting mechanism. The sander is used to sand the furniture components. The guide rack is mounted on the output end of the sander and is used to export the sanded furniture components. The translational conveying mechanism is mounted on the output end of the guide rack and is used to transport the clamping components exported from the guide rack. The positioning detection mechanism is mounted above the translational conveying mechanism. The sorting mechanism consists of a pair, respectively mounted on both sides of the translational conveying mechanism away from the guide rack, and is used to export and sort qualified and unqualified furniture components transported on the translational conveying mechanism. This equipment can automatically detect the sanded clamping components and sort the qualified and unqualified furniture components, improving processing efficiency.

[0004] However, the following problems still exist in the existing technology. Traditional task execution methods heavily rely on pre-programmed fixed paths. However, these pre-defined paths lack adaptive adjustment capabilities when faced with different types of angle grinder assembly components. For example, when handling and loading materials of different shapes, the differences in shape lead to differences in the force applied to the clamps, and the superimposed inertial force during handling may cause the handling robot arm to swing, experience sudden stress changes, or other abnormalities, potentially leading to goods falling off, affecting the monitoring of the handling robot arm's offset, and reducing positioning accuracy. Summary of the Invention

[0005] To address this, the present invention provides a computer vision-based monitoring system for the entire process of grinding machine manufacturing. This system overcomes the problems in the prior art where, when handling and loading materials of different shapes, the shape differences lead to differences in the force exerted on the clamps, and the superimposed inertial force during the handling process may cause the handling robot arm to swing, experience sudden stress changes, or other abnormalities, potentially leading to the goods falling off and affecting the monitoring of the offset of the handling robot arm, thus reducing positioning accuracy.

[0006] To achieve the above objectives, the present invention provides a computer vision-based monitoring system for the entire process of grinding machine manufacturing, comprising: The pre-simulation module is used to obtain the target path for the handling robot arm to perform the handling task; The force analysis module is used to perform handling verification and testing on various angle grinder assembly components based on the target path controlled handling robot arm. This includes handling a single type of angle grinder assembly component based on the target path, identifying several force interference segments in the target path, and obtaining the force interference characteristics of the single type of angle grinder assembly component in the force interference segments. The category determination module is used to calculate the handling force characterization value based on the force interference characteristics of a single type of angle grinder assembly component, and determine the handling force category of the single type of angle grinder assembly component. The monitoring and control module responds to the angle grinder assembly components that need to be moved, determines the force category of the angle grinder assembly components during handling, and monitors and controls the handling robot arm based on the force category. The handling parameters of the handling robot are adjusted based on the force interference characteristics. A verification space surrounding the force interference segment is constructed based on the force interference segment. The abnormality of the handling robot is detected based on the relative positional relationship between the verification space and the real-time spatial coordinates of the end of the handling robot. The force interference characteristics include the swing amplitude and stress fluctuation value, and the range of the verification space is determined based on the transport force characterization value corresponding to the force interference segment.

[0007] Furthermore, the force analysis module is used to determine several force interference segments in the target path, including: Used to determine the trajectory end point and trajectory inflection point in the target path, to determine the trajectory segment of a predetermined length before reaching the trajectory end point, and to determine the trajectory segment of a predetermined length before and after the trajectory inflection point. This is used to determine each of the aforementioned trajectory segments as force interference segments.

[0008] Furthermore, the force interference characteristics include stress fluctuation values ​​and oscillation amplitude; The stress fluctuation value is the variance of the force amplitude at the end of the handling robotic arm when the angle grinder assembly is transported to the corresponding force interference section; The swing amplitude is determined based on the real-time coordinates of the end of the handling robotic arm when the angle grinder assembly is moved to the corresponding force interference segment.

[0009] Furthermore, the category determination module calculates the handling force characterization values ​​based on the force interference characteristics of a single type of angle grinder assembly component, including: Determine the force interference characteristics corresponding to each force interference segment, and calculate the average swing amplitude and the average stress fluctuation value; The ratio of the average swing amplitude to the swing amplitude threshold is calculated as the first force interference factor; The ratio of the average stress fluctuation value to the stress fluctuation threshold is calculated as the second force interference factor; The weighted sum of the first force interference factor and the second force interference factor is used to obtain the transport force characterization value.

[0010] Furthermore, the category determination module determines the handling force category of a single type of angle grinder assembly component, including: If the handling force characterization value corresponding to the single-type angle grinder assembly component is greater than or equal to the preset handling force threshold, it is determined to be a discrete handling force category. If the handling force characterization value corresponding to the single-type angle grinder assembly component is less than the preset handling force threshold, it is determined to be a non-discrete handling force category.

[0011] Furthermore, the monitoring and control module determines the handling force category of the angle grinder assembly components, and monitors and controls the handling robotic arm based on the handling force category, wherein... If the handling force category is discrete handling force category, then monitoring and control of the handling robot arm will be performed.

[0012] Furthermore, the monitoring and control module adjusts the handling parameters of the handling robot arm based on the force interference characteristics, including: The handling speed of the robotic arm is reduced, and the reduction amount is positively correlated with the handling force characterization value.

[0013] Furthermore, the monitoring and control module constructs a verification space surrounding the force interference segment based on the force interference segment, including: A cylindrical space is generated based on the force interference segment, and the cylindrical space is determined as the verification space. The center of any cross section of the cylindrical space perpendicular to the force interference segment is a circle, and the center of each circle is on the force interference segment. The range of the verification space is positively correlated with the transport force characterization value.

[0014] Furthermore, the monitoring and control module detects whether there are any abnormalities in the handling robot arm based on the relative positional relationship between the verification space and the real-time spatial coordinates of the end effector of the handling robot arm. Determine the set of coordinates corresponding to the verification space. If the real-time spatial coordinates of the end effector of the handling robot are not within the set of coordinates, it is determined that the handling robot has an anomaly.

[0015] Furthermore, the handling task includes moving several furniture components to the corresponding target area.

[0016] Compared with the prior art, the beneficial effects of the present invention are as follows: by obtaining the target path of the handling robot arm to perform the handling task, the handling robot arm is controlled based on the target path to perform handling verification and detection for various angle grinder assembly components, the force interference segment and force interference characteristics are determined, and the handling force category of a single type of angle grinder assembly component is further determined. Subsequently, based on the handling force category of the angle grinder assembly component, the handling robot arm is monitored and controlled. The present invention adaptively adjusts the handling parameters and detection strategies to address the impact of inertia and internal shaking of the angle grinder assembly component on the handling process and monitoring accuracy when performing the required handling task, thereby ensuring the operating efficiency and monitoring accuracy of the handling robot arm and improving the stability and reliability of the angle grinder parts assembly and handling process.

[0017] In particular, this invention uses a target path-controlled handling robot arm for handling verification and monitoring to determine the force interference segment. In reality, the force interference segment is determined based on the trajectory end point and trajectory inflection point. Because the shape difference of the target object leads to different forces on the clamp, and under the influence of superimposed inertia, the force interference is not most intense at the end of acceleration / deceleration and the moment of change of direction along the entire target path. Based on this, the force interference characteristics of the force interference segment are obtained, reflecting the influence of shape difference and inertia superposition on the handling robot arm during the handling process. This provides data support for subsequent classification of handling force categories, thereby ensuring the operating efficiency of the handling robot arm, ensuring the monitoring accuracy of the handling robot arm, and improving the stability and reliability of the angle grinder parts assembly and handling process.

[0018] In particular, this invention determines the force interference characteristics of the force interference section and quantitatively collects two key features: swing amplitude and stress fluctuation value. The swing amplitude directly reflects the spatial drift of the end effector of the robotic arm and is the most intuitive manifestation of swaying. The stress fluctuation value (calculated by the variance of the force amplitude) reveals the degree of drastic change in the dynamic load borne by the internal structure of the robotic arm. The influence of the superposition relationship of the object's swaying on the handling robotic arm is quantified from two dimensions, thereby classifying the handling force categories. Subsequently, the handling robotic arm is monitored and controlled based on the handling force categories, thereby ensuring the operating efficiency of the handling robotic arm, ensuring the monitoring accuracy of the handling robotic arm, and improving the stability and reliability of the angle grinder parts assembly and handling process.

[0019] In particular, for discrete handling force categories, the severe shaking of the angle grinder assembly components and the significant dynamic interference to the handling robot arm due to superimposed inertia cause disturbances at the end of the handling robot arm, affecting its accuracy and making it prone to errors in anomaly monitoring. Therefore, an adaptive verification space is constructed with a certain allowable error, allowing for a small amount of disturbance in the handling robot arm to determine if an anomaly exists, reducing false alarms and improving the reliability of anomaly detection. Simultaneously, the handling parameters of the handling robot arm are adaptively adjusted to adapt to the corresponding type of angle grinder assembly components, reducing the impact on the handling robot arm during handling, thereby ensuring the operating efficiency and monitoring accuracy of the handling robot arm, and improving the stability and reliability of the angle grinder component assembly and handling process. Attached Figure Description

[0020] Figure 1 A schematic diagram of a cloud-based intelligent monitoring structure for the entire furniture manufacturing process, as shown in an embodiment of the invention. Figure 2 A logic block diagram for determining the handling force category of a single type of angle grinder assembly component in an embodiment of the invention; Figure 3 This is a logic block diagram illustrating the monitoring and control of a handling robotic arm based on the type of handling force, as described in an embodiment of the invention. Figure 4 This is a logic block diagram for detecting whether a handling robotic arm has an abnormality, according to an embodiment of the invention. Detailed Implementation

[0021] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0022] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0023] It should be noted that in the description of this invention, the terms "upper", "lower", "left", "right", "inner", "outer", etc., which indicate directions or positional relationships, are based on the directions or positional relationships shown in the accompanying drawings. This is only for the convenience of description and is not intended to indicate or imply that the device or element must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation of this invention.

[0024] Furthermore, it should be noted that, in the description of this invention, unless otherwise explicitly specified and limited, the term "connection" should be interpreted broadly. For example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium; it can be a connection within two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0025] Please see Figure 1 As shown, this is a schematic diagram of a cloud-based intelligent monitoring structure for the entire furniture manufacturing process according to an embodiment of the present invention. The computer vision-based full-process monitoring system for polishing machine manufacturing according to an embodiment of the present invention includes: The pre-simulation module is used to obtain the target path for the handling robot arm to perform the handling task; The force analysis module is used to perform handling verification and testing on various angle grinder assembly components based on the target path controlled handling robot arm. This includes handling a single type of angle grinder assembly component based on the target path, identifying several force interference segments in the target path, and obtaining the force interference characteristics of the single type of angle grinder assembly component in the force interference segments. The category determination module is used to calculate the handling force characterization value based on the force interference characteristics of a single type of angle grinder assembly component, and determine the handling force category of the single type of angle grinder assembly component. The monitoring and control module responds to the angle grinder assembly components that need to be moved, determines the force category of the angle grinder assembly components during handling, and monitors and controls the handling robot arm based on the force category. The handling parameters of the handling robot are adjusted based on the force interference characteristics. A verification space surrounding the force interference segment is constructed based on the force interference segment. The abnormality of the handling robot is detected based on the relative positional relationship between the verification space and the real-time spatial coordinates of the end of the handling robot. The force interference characteristics include the swing amplitude and stress fluctuation value, and the range of the verification space is determined based on the transport force characterization value corresponding to the force interference segment.

[0026] Specifically, the specific structure of the handling robotic arm is not limited. It can be a robotic arm with a gripping structure at the end. The degree of freedom and operation mode of the robotic arm can be selected by those skilled in the art according to the site environment, which will not be elaborated here.

[0027] Specifically, the target path for the required handling task is predetermined. The handling task includes moving the angle grinder assembly to the storage location. The movement path of the robotic arm end effector during the handling process can be detected to obtain the target path, which will not be elaborated further.

[0028] Specifically, there is no limitation on the classification method of angle grinder assembly components. In actual production, it is usually necessary to grasp multiple types of angle grinder assembly components. Based on this, the same angle grinder assembly components are grouped into one category for easy centralized analysis, which will not be elaborated further here.

[0029] Specifically, the force analysis module is used to determine several force interference segments in the target path, including: Used to determine the trajectory end point and trajectory inflection point in the target path, to determine the trajectory segment of a predetermined length before reaching the trajectory end point, and to determine the trajectory segment of a predetermined length before and after the trajectory inflection point. This is used to determine each of the aforementioned trajectory segments as force interference segments.

[0030] In practice, the predetermined length is determined in advance. The purpose is to select the trajectory end point and the trajectory inflection point near the trajectory, so as to facilitate the observation of the situation when the end of the handling robot moves to the corresponding trajectory segment. In practice, the predetermined length is set to 0.05 times the total length of the target path.

[0031] This invention utilizes a target path-controlled handling robot arm for handling verification and monitoring, identifying force interference segments. In practice, these segments are determined based on the trajectory end point and trajectory inflection point. Due to differences in the shape of the target object, the clamp experiences varying forces. Furthermore, under the influence of superimposed inertia, the forces are not most intense at the end of acceleration / deceleration or the moment of direction change along the entire target path. Based on this, the force interference characteristics of the force interference segments are obtained, reflecting the impact of shape differences and inertial superposition on the handling robot arm during the handling process. This provides data support for subsequent classification of handling force categories, thereby ensuring the operating efficiency and monitoring accuracy of the handling robot arm, and improving the stability and reliability of the angle grinder parts assembly and handling process.

[0032] Specifically, the force interference characteristics include stress fluctuation values ​​and oscillation amplitude; The stress fluctuation value is the variance of the force amplitude at the end of the handling robotic arm when the angle grinder assembly is transported to the corresponding force interference section; The swing amplitude is determined based on the real-time coordinates of the end of the handling robotic arm when the angle grinder assembly is moved to the corresponding force interference segment.

[0033] Specifically, there is no limitation on the method of monitoring the force amplitude at the end of the robotic arm. In practice, the end of the robotic arm is usually an end effector, such as a gripper or welding torch. Therefore, a six-dimensional force sensor can be set at the movable joint near the end of the robotic arm to monitor the force on the end of the robotic arm in multiple directions. When calculating the force amplitude, the resultant force is calculated and used as the force amplitude. This will not be elaborated further.

[0034] Specifically, there is no limitation on the method for determining the swing amplitude of the robotic arm's end effector. The swing behavior generated by the robotic arm can be captured based on the real-time coordinates of the end effector, and the swing amplitude can then be determined. This will not be elaborated further.

[0035] Specifically, the category determination module calculates the handling force characterization value based on the force interference characteristics of a single type of angle grinder assembly component, including: Determine the force interference characteristics corresponding to each force interference segment, and calculate the average swing amplitude and the average stress fluctuation value; The ratio of the average swing amplitude to the swing amplitude threshold is calculated as the first force interference factor; The ratio of the average stress fluctuation value to the stress fluctuation threshold is calculated as the second force interference factor; The weighted sum of the first force interference factor and the second force interference factor is used to obtain the transport force characterization value.

[0036] The swing amplitude threshold and stress fluctuation threshold are predetermined, wherein, Pre-determine several operation records of the handling robot under abnormal working conditions, including the force interference characteristics of several force interference segments during operation; The mean swing amplitude and the mean stress fluctuation value are calculated. The swing amplitude threshold is set as the product of the mean swing amplitude and the error coefficient, and the stress fluctuation threshold is set as the product of the mean stress fluctuation value and the error coefficient. The error coefficient is selected in the range [0.85, 0.95], preferably 0.9.

[0037] Specifically, please refer to Figure 2 As shown, Figure 2 This is a logic block diagram illustrating the determination of the handling force category of a single-type angle grinder assembly component according to an embodiment of the invention. The category determination module determines the handling force category of the single-type angle grinder assembly component by including: If the handling force characterization value corresponding to the single-type angle grinder assembly component is greater than or equal to the preset handling force threshold, it is determined to be a discrete handling force category. If the handling force characterization value corresponding to the single-type angle grinder assembly component is less than the preset handling force threshold, it is determined to be a non-discrete handling force category.

[0038] The handling force threshold is selected within the range [1.15, 1.3].

[0039] This invention identifies the force interference characteristics of the force interference segment and quantitatively collects two key features: swing amplitude and stress fluctuation value. The swing amplitude directly reflects the spatial drift of the end effector of the robotic arm and is the most intuitive manifestation of swaying. The stress fluctuation value (calculated by the variance of the force amplitude) reveals the degree of drastic change in the dynamic load borne by the internal structure of the robotic arm. The influence of the superposition relationship of the object's swaying on the handling robotic arm is quantified from two dimensions, thereby classifying the handling force categories. Subsequently, the handling robotic arm is monitored and controlled based on the handling force categories, thereby ensuring the operating efficiency and monitoring accuracy of the handling robotic arm, and improving the stability and reliability of the angle grinder parts assembly and handling process.

[0040] Specifically, please refer to Figure 3 As shown, Figure 3 This is a logic block diagram illustrating the monitoring and control of a robotic arm based on the type of handling force according to an embodiment of the invention. The monitoring and control module determines the type of handling force of the angle grinder assembly and monitors and controls the robotic arm based on this type of handling force. If the handling force category is discrete handling force category, then monitoring and control of the handling robot arm will be performed.

[0041] Specifically, the monitoring and control module adjusts the handling parameters of the handling robot arm based on the force interference characteristics, including: The handling speed of the robotic arm is reduced, and the reduction amount is positively correlated with the handling force characterization value.

[0042] In implementation, optional, The ratio of the difference between the transport stress characterization value and the transport stress threshold is used as the speed adjustment coefficient; The product of the initial speed and the speed adjustment coefficient is determined as the reduction in transport speed.

[0043] Understandably, the transport speed is the moving speed of the robotic arm after it picks up the angle grinder assembly components, and the difference ratio is the ratio of the difference between two values ​​to the mean of the two values.

[0044] Specifically, the monitoring and control module constructs a verification space surrounding the force interference segment based on the force interference segment, including: A cylindrical space is generated based on the force interference segment, and the cylindrical space is determined as the verification space. The center of any cross section of the cylindrical space perpendicular to the force interference segment is a circle, and the center of each circle is on the force interference segment. The range of the verification space is positively correlated with the transport force characterization value.

[0045] It is understandable that a circle can be created at the beginning of the force interference segment, and the circle can be moved along the force interference segment. During the movement, the circle is always perpendicular to the force interference segment, and thus the corresponding verification space can be obtained. Of course, other methods can also be used, as long as they can meet the corresponding requirements.

[0046] In implementation, optional, The range of the verification space can be defined by adjusting the diameter of the cylindrical space section. The larger the diameter, the larger the range. In practice, the ratio of the handling force characterization value to the handling force threshold can be calculated as the range adjustment coefficient. The product of the reference diameter and the range adjustment factor is calculated to obtain the adjusted diameter, and then the range of the verification space is determined. In practice, the volume of the verification space can be regarded as the range, which will not be elaborated further.

[0047] The reference diameter is determined based on the swing amplitude threshold and is set to 1.1 times the swing amplitude threshold.

[0048] Specifically, please refer to Figure 4 As shown, Figure 4 This is a logic block diagram illustrating the detection of abnormalities in a robotic arm according to an embodiment of the invention. The monitoring and control module detects abnormalities in the robotic arm based on the relative positional relationship between the verification space and the real-time spatial coordinates of the robotic arm's end effector. Determine the set of coordinates corresponding to the verification space. If the real-time spatial coordinates of the end effector of the handling robot are not within the set of coordinates, it is determined that the handling robot has an anomaly.

[0049] The moving task includes moving several furniture parts to the corresponding target area.

[0050] For discrete handling force categories, the severe shaking of the angle grinder assembly components and the significant dynamic interference to the handling robot arm due to superimposed inertia cause end-effector disturbances, affecting the robot arm's accuracy and making it prone to errors in anomaly detection. Therefore, an adaptive verification space is constructed with a certain allowable error, allowing for a small amount of disturbance in the handling robot arm to determine if an anomaly exists, reducing false triggers and improving the reliability of anomaly detection. Simultaneously, the handling parameters of the robot arm are adaptively adjusted to adapt to the corresponding type of angle grinder assembly components, reducing the impact on the robot arm during handling, thereby ensuring the robot arm's operating efficiency, monitoring accuracy, and improving the stability and reliability of the angle grinder component assembly and handling process.

[0051] A system for monitoring the entire grinding machine manufacturing process based on computer vision is also provided, comprising: The task perception module is used to obtain the target path of the handling robot arm to perform the handling task, and to control the handling robot arm to perform handling verification and detection based on the target path for various angle grinder assembly components. The classification module is used to calculate the handling force characterization value based on the force interference characteristics of a single type of angle grinder assembly component, and determine the handling force category of the single type of angle grinder assembly component. The control module, in response to the angle grinder assembly components to be moved, determines the type of force applied to the angle grinder assembly components during handling and monitors and controls the handling robot arm based on the type of force applied. The handling parameters of the handling robot are adjusted based on the force interference characteristics. A verification space surrounding the force interference segment is constructed based on the force interference segment. The abnormality of the handling robot is detected based on the relative positional relationship between the verification space and the real-time spatial coordinates of the end of the handling robot.

[0052] Specifically, there are no restrictions on the structure of the task perception module, classification module, and control module; they can all be composed of logic components, including field-programmable processors, computers, or microprocessors in computers.

[0053] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. A computer vision-based monitoring system for the entire process of grinding machine manufacturing, characterized in that, The pre-simulation module is used to obtain the target path for the handling robot arm to perform the handling task; The force analysis module is used to perform handling verification and testing on various angle grinder assembly components based on the target path controlled handling robot arm. This includes handling a single type of angle grinder assembly component based on the target path, identifying several force interference segments in the target path, and obtaining the force interference characteristics of the single type of angle grinder assembly component in the force interference segments. The category determination module is used to calculate the handling force characterization value based on the force interference characteristics of a single type of angle grinder assembly component, and determine the handling force category of the single type of angle grinder assembly component. The monitoring and control module responds to the angle grinder assembly components that need to be moved, determines the force category of the angle grinder assembly components during handling, and monitors and controls the handling robot arm based on the force category. The handling parameters of the handling robot are adjusted based on the force interference characteristics. A verification space surrounding the force interference segment is constructed based on the force interference segment. The abnormality of the handling robot is detected based on the relative positional relationship between the verification space and the real-time spatial coordinates of the end of the handling robot. The force interference characteristics include the swing amplitude and stress fluctuation value, and the range of the verification space is determined based on the transport force characterization value corresponding to the force interference segment.

2. The computer vision-based full-process monitoring system for grinding machine manufacturing as described in claim 1, characterized in that, The force analysis module is used to determine several force interference segments in the target path, including: Used to determine the trajectory end point and trajectory inflection point in the target path, to determine the trajectory segment of a predetermined length before reaching the trajectory end point, and to determine the trajectory segment of a predetermined length before and after the trajectory inflection point. This is used to determine each of the aforementioned trajectory segments as force interference segments.

3. The computer vision-based full-process monitoring system for grinding machine manufacturing as described in claim 2, characterized in that, The force interference characteristics include stress fluctuation values ​​and oscillation amplitude; The stress fluctuation value is the variance of the force amplitude at the end of the handling robotic arm when the angle grinder assembly is transported to the corresponding force interference section; The swing amplitude is determined based on the real-time coordinates of the end of the handling robotic arm when the angle grinder assembly is moved to the corresponding force interference segment.

4. The computer vision-based full-process monitoring system for grinding machine manufacturing as described in claim 1, characterized in that, The category determination module calculates the handling force characterization values ​​based on the force interference characteristics of a single type of angle grinder assembly component, including: Determine the force interference characteristics corresponding to each force interference segment, and calculate the average swing amplitude and the average stress fluctuation value; The ratio of the average swing amplitude to the swing amplitude threshold is calculated as the first force interference factor; The ratio of the average stress fluctuation value to the stress fluctuation threshold is calculated as the second force interference factor; The weighted sum of the first force interference factor and the second force interference factor is used to obtain the transport force characterization value.

5. The computer vision-based full-process monitoring system for grinding machine manufacturing according to claim 4, characterized in that, The category determination module determines the handling force category of a single type of angle grinder assembly component, including: If the handling force characterization value corresponding to the single-type angle grinder assembly component is greater than or equal to the preset handling force threshold, it is determined to be a discrete handling force category. If the handling force characterization value corresponding to the single-type angle grinder assembly component is less than the preset handling force threshold, it is determined to be a non-discrete handling force category.

6. The computer vision-based full-process monitoring system for grinding machine manufacturing according to claim 1, characterized in that, The monitoring and control module determines the force category of the angle grinder assembly components during handling, and monitors and controls the handling robotic arm based on the force category. If the handling force category is discrete handling force category, then monitoring and control of the handling robot arm will be performed.

7. The computer vision-based full-process monitoring system for grinding machine manufacturing according to claim 1, characterized in that, The monitoring and control module adjusts the handling parameters of the handling robot arm based on the force interference characteristics, including: The handling speed of the robotic arm is reduced, and the reduction amount is positively correlated with the handling force characterization value.

8. The computer vision-based full-process monitoring system for grinding machine manufacturing according to claim 1, characterized in that, The monitoring and control module constructs a verification space surrounding the force interference segment based on the force interference segment, including: A cylindrical space is generated based on the force interference segment, and the cylindrical space is determined as the verification space. The center of any cross section of the cylindrical space perpendicular to the force interference segment is a circle, and the center of each circle is on the force interference segment. The range of the verification space is positively correlated with the transport force characterization value.

9. The computer vision-based full-process monitoring system for grinding machine manufacturing according to claim 1, characterized in that, The monitoring and control module detects whether there are any abnormalities in the handling robot arm based on the relative positional relationship between the verification space and the real-time spatial coordinates of the end effector. Determine the set of coordinates corresponding to the verification space. If the real-time spatial coordinates of the end effector of the handling robot are not within the set of coordinates, it is determined that the handling robot has an anomaly.

10. The computer vision-based full-process monitoring system for grinding machine manufacturing according to claim 1, characterized in that, The moving task includes moving several furniture parts to the corresponding target area.

Images