A grinding trajectory control method and system for automatic grinding of wheel hub burrs

Abstract

The present invention discloses a grinding trajectory control method and system for automatic grinding of wheel hub burrs, which relates to the field of electronic engineering technology, including scanning the wheel hub through visual detection, analyzing its material, capturing its three-dimensional model and locating all burrs; analyzing burr features through DBSCAN algorithm and generating corresponding grinding paths; the robot arm starts grinding according to the grinding path and collects data in real time during the grinding process; continuously detects burrs on the wheel hub surface through visual sensors, and the robot triggers an alarm mechanism after finding burrs, and adjusts the grinding path according to the specific location of the burrs, and grinds the wheel hub again until the detection is qualified. The present invention realizes comprehensive wheel hub surface data collection by starting an industrial camera, generates an optimal grinding path to achieve optimization of path selection, realizes immediate response and processing of abnormal situations through alarm mechanism and path adjustment, and realizes closed-loop control and high-quality assurance of the grinding process through repeated verification and adjustment.

Description

Technical Field

[0001] The invention relates to the technical field of electronic engineering, and in particular to a grinding track control method and system for automatic grinding of wheel hub burrs. Background Art

[0002] In recent years, with the rapid development of the automotive industry and the advancement of automation technology, the wheel manufacturing process has also been significantly improved. Traditional wheel grinding mainly relies on manual operation, which is not only inefficient but also difficult to ensure consistent quality. In order to solve these problems, automatic grinding technology has gradually become a research hotspot. Existing automatic grinding technology mostly uses preset paths for grinding. Although it can improve production efficiency to a certain extent, its flexibility and adaptability are still insufficient when facing complex-shaped wheels and surface defects. Especially when dealing with non-standard and special-shaped wheels, traditional methods often cannot achieve accurate burr positioning and efficient removal.

[0003] In the existing technology, although some grinding technologies have introduced visual detection methods to preliminarily identify the burr location, there are many limitations in practical applications. Most grinding technologies can only provide static burr distribution information and lack effective monitoring of dynamic changes during the grinding process. For wheels with different materials and complex geometric structures, existing algorithms are difficult to accurately capture and analyze burr characteristics, and their ability to generate the optimal grinding path is limited. Summary of the invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a grinding trajectory control method for automatic grinding of wheel hub burrs, which solves the problems of low efficiency and difficulty in ensuring consistent quality, lack of flexibility and adaptability in handling surface defects, inability to achieve accurate burr positioning and efficient removal, and also solves the lack of monitoring of dynamic changes during the grinding process, and the difficulty for the algorithm to accurately capture and analyze burr characteristics.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, the present invention provides a grinding trajectory control method for automatic grinding of wheel hub burrs, which comprises scanning the wheel hub by visual inspection, analyzing its material, capturing its three-dimensional model and locating all burrs;

[0008] Analyze burr features through DBSCAN algorithm and generate corresponding grinding paths;

[0009] The robot arm starts grinding according to the grinding path and collects data in real time during the grinding process;

[0010] The wheel hub surface is continuously inspected for burrs through visual sensors. When a burr is found, the robot triggers the alarm mechanism and adjusts the grinding path according to the specific location of the burr. The wheel hub is then polished again until it passes the inspection.

[0011] As a preferred solution of the grinding trajectory control method for automatic grinding of wheel hub burrs of the present invention, the method of scanning the wheel hub by visual inspection, analyzing its material, capturing its three-dimensional model and locating all burrs includes the following steps:

[0012] Start the industrial camera to take the first photo of the front of the wheel hub to obtain the 3D model, material information, burr location and features of the entire wheel hub surface;

[0013] After the shooting is completed, switch to multispectral imaging mode to collect image data of the wheel hub from multiple wavelength ranges;

[0014] Smoothing the collected wheel hub image data to remove noise points and outliers;

[0015] The 3D model of the hub surface, material information, burr location and processed hub image data are integrated to form a comprehensive data set.

[0016] As a preferred solution of the grinding trajectory control method for automatic grinding of wheel hub burrs according to the present invention, wherein: analyzing burr features by DBSCAN algorithm and generating cluster labels includes the following steps:

[0017] Extracting burr features from the locations of burrs in the comprehensive dataset;

[0018] Mark all burr features as unvisited. For each unvisited data point Z, check the number of points in its neighborhood. When the number of points in the neighborhood is less than Z, Z is marked as a noise point. When the number of points in the neighborhood is greater than or equal to Z, Z is considered to be a core point and a new cluster is started. For each core point Z, add all points in its neighborhood to the current cluster and continue to expand the cluster until no new points can be added.

[0019] The core point gets a unique cluster label according to the cluster it belongs to.

[0020] As a preferred solution of the grinding trajectory control method for automatic grinding of wheel hub burrs according to the present invention, generating a corresponding grinding path based on cluster labels includes the following steps:

[0021] According to the cluster label, the importance score expression of each glitch is calculated as;

[0022] ;

[0023] in, represents the glitch index variable, is the glitch importance score, is the cluster label of the burr, is the neighborhood radius, Represents the central range. It is a burr In the neighborhood The points within is the weight coefficient, It is a burr point Characteristics and Gaussian components, is the normalization constant, is the number of components of the Gaussian mixture model, is the index variable of the Gaussian mixture model, The probability density of the Gaussian distribution, It is The mean vector of the Gaussian components;

[0024] The importance scores of each glitch are combined into a list, and a sorting algorithm is used to sort all the glitches in the list from high to low according to the scores to obtain a sorted glitch list;

[0025] The sorted burr list is prioritized from the burr with the highest score to generate the optimal grinding path, the expression is;

[0026] ;

[0027] in, Is the optimal grinding path, is the index of the path segment, is the number of path segments, Is a path segment The cost, is the normalization factor, is the first Points, is the first Points, Is a path segment The Euclidean distance of Is a path segment The entropy weight of is the glitch importance score, Is a path segment The entropy value of is the path stability coefficient.

[0028] As a preferred solution of the grinding trajectory control method for automatic grinding of wheel hub burrs according to the present invention, wherein: based on the grinding path, the robot arm starts grinding according to the grinding path, and real-time collection of data during the grinding process includes the following steps:

[0029] Load the generated grinding path into the robot's control panel and initialize all sensors;

[0030] The robot arm starts grinding according to the generated grinding path, collects contact force, surface texture changes and vibration during the grinding process from various sensors, and continuously monitors the changes in the wheel surface.

[0031] As a preferred solution of the grinding trajectory control method for automatic grinding of wheel hub burrs according to the present invention, the burr detection on the wheel hub surface and the adjustment of the grinding path include the following steps:

[0032] The wheel hub surface is monitored by visual sensors, the detected burrs are classified, and new burrs and burrs that have been processed but not completely removed are distinguished. The robot triggers the alarm mechanism as follows:

[0033] ;

[0034] in, is the abnormal glitch detected, is the cluster label of the burr, When it exceeds the normal range, the robot enters the alarm state. is within the normal range, the robot's status, It's time. It's time The actual measured value, is the Kalman filter prediction value, is the standard deviation of the measurement error, is the alarm threshold, is the starting point of the time interval. is the end point of the time interval;

[0035] The alarm notifies the operator through sound and light signals, and records the time and location of the abnormal event, and adjusts the grinding path according to the specific location of the burr. The expression is:

[0036] ;

[0037] in, is the adjusted grinding path, is the discrete time point index, is the number of discrete time points on the path, is the polishing parameter index, is the regularization coefficient, is the number of grinding parameters, is the grinding parameter, is the starting point of the time interval. is the end of the time interval, It is discrete time points.

[0038] As a preferred solution of the grinding trajectory control method for automatic grinding of wheel hub burrs of the present invention, the re-grinding until the test is qualified includes the following steps:

[0039] Control the robot arm to grind again according to the adjusted grinding path;

[0040] After grinding, the wheel hub surface is rechecked through the visual sensor. If the visual sensor fails the detection, the alarm mode is triggered again, and the grinding path is adjusted again according to the specific location of the burr for grinding. If the visual sensor passes the detection, the grinding work is completed.

[0041] In a second aspect, the present invention provides a grinding trajectory control system for automatic grinding of wheel hub burrs, comprising a visual detection module, a path generation module, a grinding module, and an alarm and adjustment module;

[0042] The visual inspection module is used to scan the wheel hub through visual inspection, analyze its material, capture its three-dimensional model and locate all burrs;

[0043] The path generation module is used to analyze the burr features through the DBSCAN algorithm and generate a corresponding grinding path;

[0044] The polishing module is used for the robot arm to start polishing according to the polishing path and collect data during the polishing process in real time;

[0045] The alarm and adjustment module is used to continuously detect burrs on the surface of the wheel hub through a visual sensor. When a burr is found, the robot triggers an alarm mechanism, adjusts the grinding path according to the specific location of the burr, and grinds the wheel hub again until it passes the inspection.

[0046] The beneficial effects of the present invention are as follows: by starting a high-resolution 3D scanner, comprehensive wheel hub surface data collection is achieved, data integration realizes unified management and application of multi-source information, the DBSCAN algorithm is used to analyze burr features and generate cluster labels, intelligent classification and grouping of burr features are achieved, the importance score of burrs is calculated to achieve quantitative evaluation of burr priorities, the sorting algorithm generates a burr processing priority list to achieve the optimal arrangement of the grinding sequence, the optimal grinding path is generated to achieve the optimization of path selection, the alarm mechanism and path adjustment realize immediate response and processing of abnormal situations, and repeated verification and adjustment realize closed-loop control and high quality assurance of the grinding process. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0048] Figure 1 This is a flow chart of the grinding trajectory control method for automatic grinding of wheel hub burrs in Example 1.

[0049] Figure 2 This is a module diagram of the grinding trajectory control system for automatic grinding of wheel hub burrs in Example 1. DETAILED DESCRIPTION

[0050] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0051] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0052] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0053] Example 1, reference Figure 1 and Figure 2 , which is the first embodiment of the present invention, provides a grinding trajectory control method for automatic grinding of wheel hub burrs, comprising the following steps:

[0054] S1. Scan the wheel hub through visual inspection to analyze its material, capture its 3D model and locate all burrs as follows:

[0055] Start the industrial camera and take the first photo of the front of the wheel hub. The industrial camera is mounted on the robot, which can be the arm, head or other position. During the shooting process, the robot automatically adjusts the shooting angle and distance to obtain high-quality original images. The collected original images are processed by edge detection and morphological operations. Gaussian filtering and median filtering algorithms are used to smooth the image to reduce noise interference. The box plot method is used to detect and remove outliers in the image, identify and mark the burr position, record the position, size and shape of each burr, and integrate the three-dimensional model of the wheel hub surface, material information and processed image data into a comprehensive data set.

[0056] Furthermore, three-dimensional data is a data set that describes the geometric shape, position, and size of the wheel hub on the X-axis, Y-axis, and Z-axis.

[0057] S2. Analyze the burr features through the DBSCAN algorithm and generate the corresponding grinding path, as follows:

[0058] Load the comprehensive data set into the robot, extract all the locations marked as burrs from the comprehensive data set to form a set of burr points. Construct a feature vector for each burr point, normalize the feature vector to ensure that the numerical ranges between different features are consistent, avoid the features from having too much impact on the clustering results, and mark all burr features as unvisited. For each unvisited data point Z, check the number of points in its neighborhood. When the number of points in the neighborhood is less than Z, Z is marked as a noise point. When the number of points in the neighborhood is greater than or equal to Z, Z is considered to be a core point and a new cluster is started. For each core point Z, add all the points in its neighborhood to the current cluster, and continue to expand the cluster until no new points can be added. Create a label counter with an initial value of 0. Assign a unique cluster label to each newly formed cluster, that is, use the current value of the label counter as the cluster label. According to the cluster label, calculate the importance score expression of each burr.

[0059] ;

[0060] in, represents the glitch index variable, which is used to identify different glitch points. is the importance score of the glitch, indicating the The importance of burr points, is the cluster label of the burr, containing the features of the burr point, is the neighborhood radius, which defines the range of each burr point considered in its surrounding space, It represents the center range, indicating how many burr points there are within the range. It is a burr In the neighborhood The number of points within is used for distance Less than or equal to How many other burr points are there within the range? is a normalization constant used to adjust the impact on the score to ensure that the score value is within a reasonable range. is the weight coefficient used for each Gaussian component The relative importance of is the index variable of the Gaussian mixture model, which is used to traverse each Gaussian component. is the number of components of the Gaussian mixture model, the number of Gaussian components used to model the data distribution, is the probability density of the Gaussian distribution with mean and the covariance matrix is a Gaussian distribution with parameters The probability density value at .

[0061] Create an empty list to store all burr points and their corresponding importance scores, traverse all burr points, and store their index and calculated importance score as a pair of tuples in the created empty list to ensure that all burr points are included in the list and each burr point has a corresponding score value. Select the quick sort algorithm to sort the burr points in descending order according to the importance score, that is, the burr with the highest score is at the front. The sorted burr list prioritizes the burr with the highest score to generate the optimal grinding path. The expression is;

[0062] ;

[0063] in, is the optimal grinding path, indicating the final selected grinding path. is the index of the path segment, used to identify each segment in the path, is the number of path segments, indicating how many segments the entire polishing path consists of. Is a path segment The cost reflects the resource consumption required to execute the path segment. is a normalization factor used to adjust the influence of distance. is the first points, representing path segments The distance is the first points, representing path segments The distance Is a path segment The entropy weight reflects the complexity and uncertainty of the path segment. Is a path segment The entropy value of , which measures the uncertainty or complexity of the path segment, is the path stability coefficient, which is used to adjust the stability of the entire path.

[0064] S3, the robot arm starts grinding according to the grinding path, and collects data in real time during the grinding process, as follows:

[0065] The calculated optimal grinding path is transmitted to the robot control panel through the network. The path data should include the coordinates, speed and acceleration information of each path point. Start the robot control panel and enter the path loading mode. Select and load the generated grinding path file from the network server. Make sure that all sensors are correctly connected to the robot control panel. Calibrate each sensor to ensure its measurement accuracy and stability. Set the status of all sensors to the initial state and be ready to receive data. Send a start command through the control panel to command the robot arm to start grinding according to the loaded grinding path. The robot arm moves accurately according to the coordinates and speed instructions of the path points to ensure that each path segment is strictly executed according to the predetermined trajectory. The contact force sensor monitors the force applied during the grinding process in real time to ensure that the wheel hub is not damaged. The surface texture sensor continuously scans the wheel hub surface during the grinding process to capture any subtle changes. The vibration sensor continuously monitors the vibration during the grinding process to detect whether there is abnormal vibration. Through high-resolution cameras and other visual sensors, the changes on the wheel hub surface are monitored in real time to ensure that the grinding process proceeds smoothly.

[0066] S4. Continuously detect burrs on the wheel hub surface through visual sensors. When burrs are found, the robot triggers the alarm mechanism and adjusts the grinding path according to the specific location of the burrs. The wheel hub is polished again until it passes the test. The details are as follows:

[0067] Start the visual sensor, and use it to continuously detect burrs on the hub surface. The robot can monitor the presence and changes of burrs in real time during the grinding process, collect images of the hub surface, and transmit them to the image processing unit to perform denoising and contrast enhancement operations on the image to ensure that the image is clear and recognizable. Compare the current image with the previous grinding record, identify the newly appeared burrs and the burrs that have been processed but not completely removed, mark the specific location of each burr in the image, and assign the corresponding cluster label, set the alarm threshold according to the size, shape and number of the burrs, sound an alarm through the buzzer to alert the operator, and display a red warning signal through the indicator light and display to further attract the operator's attention. The timestamp expression for recording each alarm is:

[0068] ;

[0069] in, It is the abnormal burr detected, which is used to indicate whether the alarm status is "triggered" or "normal". The alarm state is "triggered" when the cumulative error sum exceeds the threshold When the system triggers an alarm, The alarm state is "normal" when the cumulative error sum of squares does not exceed the threshold The system thinks everything is normal. is the starting point of the time interval. is the end point of the time interval, defining the time interval of integration for evaluating changes in the glitch processing process. is the actual measured value used in time The actual measurement data at is the Kalman filter prediction value for time The predicted or expected measured value at is the standard deviation of the measurement error, which is used to measure the degree of fluctuation of the data. It is the alarm threshold, which is used to determine whether the cumulative error sum of squares exceeds the allowable range.

[0070] The grinding path expression can be adjusted according to the specific location of the burr:

[0071] ;

[0072] in, is the adjusted grinding path to ensure a smooth and efficient path. is the time point index used to identify each discrete time point in the path , is the number of discrete time points on the path, which is the total number of time points considered during the entire polishing process, is a discrete time point index, used to identify each discrete time point in the path, is the regularization coefficient, which is used to adjust the relative importance between different terms. is the number of polishing parameters, which are used as constraints in path optimization. is the grinding parameter used to ensure its stability. is a variable index used to identify multiple constraints involved in path optimization. is the starting point of the time interval. is the end point of the time interval and is used to evaluate the smoothness of the path over the entire time period. It is discrete time points, used to identify each discrete time point in the path.

[0073] This embodiment also provides a grinding trajectory control system for automatic grinding of wheel hub burrs, including: a visual detection module, a path generation module, a grinding module, and an alarm and adjustment module;

[0074] The visual inspection module is used to scan the wheel hub through visual inspection, analyze its material, capture its three-dimensional model and locate all burrs; the path generation module is used to analyze the burr characteristics through the DBSCAN algorithm and generate the corresponding grinding path; the grinding module is used for the robot arm to start grinding according to the grinding path and collect data during the grinding process in real time; the alarm and adjustment module is used to continuously detect burrs on the surface of the wheel hub through visual sensors. When a burr is found, the robot triggers the alarm mechanism and adjusts the grinding path according to the specific location of the burr, and grinds the wheel hub again until it passes the inspection.

[0075] In summary, the present invention realizes comprehensive wheel hub surface data collection by starting a high-resolution 3D scanner, integrates data to realize unified management and application of multi-source information, applies the DBSCAN algorithm to analyze burr features and generate cluster labels, realizes intelligent classification and grouping of burr features, calculates the importance score of burrs to realize quantitative evaluation of burr priorities, generates a burr processing priority list using a sorting algorithm to realize the optimal arrangement of the grinding order, generates the optimal grinding path to realize the optimization of path selection, uses an alarm mechanism and path adjustment to realize immediate response and processing of abnormal situations, and repeatedly verifies and adjusts to realize closed-loop control and high-quality assurance of the grinding process.

[0076] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A grinding trajectory control method for automatic grinding of wheel hub burrs, characterized in that: include, Scan the wheel hub through visual inspection to analyze its material, capture its 3D model and locate any burrs; Analyze burr features through DBSCAN algorithm and generate corresponding grinding paths; The robot arm starts grinding according to the grinding path and collects data in real time during the grinding process; The wheel hub surface is continuously inspected for burrs using visual sensors. When a burr is detected, the robot triggers an alarm mechanism and adjusts the grinding path according to the specific location of the burr. The wheel hub is then re-grinded until it passes the inspection. Based on the cluster labels, generating the corresponding polishing path includes the following steps: According to the cluster label, the importance score expression of each glitch is calculated as; Among them, i represents the burr index variable, P i is the importance score of the glitch, f i is the cluster label of the burr, ∈ is the neighborhood radius, N represents the center range, N ∈ (f i ) is the burr f i The number of points in the neighborhood ∈, w k is the weight coefficient, φ k (f i ) is the burr point f i The features and the k-th Gaussian component, γ is the normalization constant, K is the number of components of the Gaussian mixture model, k is the index variable of the Gaussian mixture model, Probability density of Gaussian distribution, μ k is the mean vector of the kth Gaussian component; The importance scores of each glitch are combined into a list, and a sorting algorithm is used to sort all the glitches in the list from high to low according to the scores to obtain a sorted glitch list; The sorted burr list is prioritized from the burr with the highest score to generate the optimal grinding path, the expression is; Where R is the optimal polishing path, r is the index of the path segment, n is the number of path segments, c r is the cost of path segment r, δ is the normalization factor, R r-1 is the r-1th point on the path, R r is the rth point on the path, d(R r-1 R r ) is the Euclidean distance of path segment r, λ r is the entropy weight of path segment r, P i is the importance score of the glitch, H(P i ) is the entropy value of path segment r, and η is the path stability coefficient.

2. The grinding trajectory control method for automatic grinding of wheel hub burrs according to claim 1, characterized in that: The method of scanning the wheel hub by visual inspection, analyzing its material, capturing its three-dimensional model and locating all burrs includes the following steps: Start the industrial camera to take the first photo of the front of the wheel hub to obtain the 3D model, material information, burr location and features of the entire wheel hub surface; After the shooting is completed, switch to multispectral imaging mode to collect image data of the wheel hub from multiple wavelength ranges; Smoothing the collected wheel hub image data to remove noise points and outliers; The 3D model of the hub surface, material information, burr location and processed hub image data are integrated to form a comprehensive data set.

3. The grinding trajectory control method for automatic grinding of wheel hub burrs according to claim 2, characterized in that: Analyzing burr features and generating cluster labels through DBSCAN algorithm includes the following steps: Extracting burr features from the locations of burrs in the comprehensive dataset; Mark all burr features as unvisited. For each unvisited data point Z, check the number of points in its neighborhood. When the number of points in the neighborhood is less than Z, Z is marked as a noise point. When the number of points in the neighborhood is greater than or equal to Z, Z is considered to be a core point and a new cluster is started. For each core point Z, add all points in its neighborhood to the current cluster and continue to expand the cluster until no new points are added. The core point gets a unique cluster label according to the cluster it belongs to.

4. The grinding trajectory control method for automatic grinding of wheel hub burrs according to claim 3, characterized in that: Based on the grinding path, the robot arm starts grinding according to the grinding path. Real-time data collection during the grinding process includes the following steps: Load the generated grinding path into the robot's control panel and initialize all sensors; The robot arm starts grinding according to the generated grinding path, collects contact force, surface texture changes and vibration during the grinding process from various sensors, and continuously monitors the changes in the wheel surface.

5. The grinding trajectory control method for automatic grinding of wheel hub burrs according to claim 4, characterized in that: The following steps are involved in detecting burrs on the hub surface and adjusting the grinding path: The wheel hub surface is monitored by visual sensors, the detected burrs are classified, and new burrs and burrs that have been processed but not completely removed are distinguished. The robot triggers the alarm mechanism as follows: Among them, A(f i ) is the abnormal glitch detected, f i is the cluster label of the glitch, Y is when the robot enters the alarm state when it exceeds the normal range, G is the state of the robot within the normal range, t is the time, y t is the actual measured value at time t, is the Kalman filter prediction value, σ t is the standard deviation of the measurement error, τ is the alarm threshold, t0 is the starting point of the time interval, t f is the end point of the time interval; The alarm notifies the operator through sound and light signals, and records the time and location of the abnormal event, and adjusts the grinding path according to the specific location of the burr. The expression is: Where R′ is the adjusted polishing path, l is the discrete time point index, D is the number of discrete time points on the path, j is the polishing parameter index, γ is the regularization coefficient, M is the number of polishing parameters, and q j is the polishing parameter, t0 is the starting point of the time interval, t f is the end point of the time interval, t l is the lth discrete time point.

6. The grinding trajectory control method for automatic grinding of wheel hub burrs according to claim 5, characterized in that: The re-grinding until the test is qualified includes the following steps: Control the robot arm to grind again according to the adjusted grinding path; After grinding, the wheel hub surface is rechecked through the visual sensor. If the visual sensor fails the detection, the alarm mode is triggered again, and the grinding path is adjusted again according to the specific location of the burr for grinding. If the visual sensor passes the detection, the grinding work is completed.

7. A grinding trajectory control system for automatic grinding of wheel hub burrs, based on the grinding trajectory control method for automatic grinding of wheel hub burrs according to any one of claims 1 to 6, characterized in that: Including visual inspection module, path generation module, polishing module and alarm and adjustment module; The visual inspection module is used to scan the wheel hub through visual inspection, analyze its material, capture its three-dimensional model and locate all burrs; The path generation module is used to analyze the burr features through the DBSCAN algorithm and generate a corresponding grinding path; The polishing module is used for the robot arm to start polishing according to the polishing path and collect data during the polishing process in real time; The alarm and adjustment module is used to continuously detect burrs on the surface of the wheel hub through a visual sensor. When a burr is found, the robot triggers an alarm mechanism, adjusts the grinding path according to the specific location of the burr, and grinds the wheel hub again until it passes the inspection.

Images