Flexible control method of machining center manufacturing process based on motion capture
By adopting motion capture-based technology in the machining center, combining laser sensors and visual analysis chips, the combination of gesture control and CNC system is realized, and temporary control instructions are generated, which solves the problems of flexibility and accurate control in the existing technology and improves processing Center adaptability and safety.
Patent Information
- Application Number
- CN202411259915.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-10
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2044-09-10
AI Technical Summary
When facing diverse machining tasks and changing process requirements, the existing intelligent control system of machining centers lacks sufficient adaptability and accuracy, especially in the combination of gesture control and CNC systems, it is difficult to achieve flexible and accurate machining processes. control.
Using a machining center manufacturing process based on motion capture, the gesture controller detects the operator's gesture instructions, uses the gesture command analysis module combined with laser sensor and visual analysis chip to generate temporary control instructions, and confirm it through the touch display to achieve flexible control. , avoid misoperation and misidentification.
It realizes flexible adjustment and precise control of the processing process, improves the adaptability and safety of the system, avoids production accidents caused by misidentification, and is suitable for noisy application environments in machining centers.
Smart Images

Figure CN118927013A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of machining center program control, and in particular to a flexible control method of a machining center manufacturing process based on motion capture. Background Art
[0002] Intelligent control of machining centers refers to the use of advanced control theory, computer technology, sensor technology and artificial intelligence algorithms to achieve automation, precision and efficiency of the machining process. The following is the current status and defects of intelligent control of machining centers:
[0003] Machining centers generally use CNC systems to automate the machining process. Intelligent programming systems can automatically generate machining programs based on CAD models, reducing the workload of manual programming. Intelligent control is highly dependent on the quality and quantity of data. Insufficient or low-quality data will affect system performance. Faced with diverse machining tasks and ever-changing process requirements, the adaptability of intelligent systems needs to be improved.
[0004] Gesture control technology is an advanced human-computer interaction method that allows operators to control the execution of objects through hand movements, improving the intuitiveness and efficiency of operations.
[0005] Therefore, how to combine gesture control technology with CNC system so that the operator can use gesture intervention to control the processing more delicately and flexibly while observing the workpiece is a technical problem that needs to be solved urgently. Summary of the invention
[0006] In order to solve the technical problem of combining gesture control technology with numerical control system, the present invention provides a flexible control method of machining center manufacturing process based on motion capture. The following technical solutions are adopted:
[0007] A flexible control method for a machining center manufacturing process based on motion capture, a gesture switch module of a gesture controller detects a front gesture command switch signal, if there is a gesture intervention command signal, the gesture controller uses the following steps to start recognizing and analyzing the gesture control signal:
[0008] Step 1: The motion picture capture module of the gesture controller first collects the operator's gesture picture, and the laser sensor module then senses the gesture scale data;
[0009] Step 2: Pack the gesture picture data and gesture scale data and send them to a gesture command analysis module based on a combination of a visual analysis chip and a data analysis chip;
[0010] Step 3: The gesture command analysis module parses the gesture screen data and gesture scale data. The visual analysis chip performs a matching degree analysis on the gesture screen data based on the standard gesture feature database stored in the memory. If the matching degree is greater than the set matching degree threshold, the corresponding control command template data stored in the memory is extracted based on the standard gesture feature, and the corresponding control command template data is transmitted to the command generation chip.
[0011] Step 4: The data analysis chip parses the gesture scale data, analyzes the control scale data contained in the gesture scale data, and transmits the control scale data to the instruction generation chip;
[0012] Step 5, the instruction generation chip combines the control scale data with the corresponding control instruction template data to generate a temporary control instruction for controlling the numerical control system;
[0013] Step 6: The temporary control instruction displays a text description corresponding to the temporary control instruction on the gesture instruction touch display, and provides an option to modify the scale data and an instruction confirmation item;
[0014] Step 7, the operator confirms the operation command through the command confirmation item on the gesture command touch display, or confirms the temporary control command after fine-tuning through the option of modifying the scale data;
[0015] Step 8: The numerical control system replaces the corresponding part of the existing machining control instruction with the temporary control instruction and executes it.
[0016] By adopting the above technical solution, a gesture switch module is used to control the entire gesture detection on and off to avoid misoperation. Specifically, it can be a gesture command switch mode, a press switch mode, or an induction switch mode;
[0017] When the gesture switch module detects the front gesture command switch signal, the action picture capture module and the laser sensor module, the action picture capture module first collects the operator's gesture picture, and the laser sensor module then senses the gesture scale data;
[0018] The operator's gesture screen is for interactive operation items, such as feed speed, program selection control, feed amount adjustment, etc.; gesture scale data can realize direct adjustment of control values, solving the problem that gesture control alone cannot accurately adjust specific values. For example, the specific value of feed amount can be directly adjusted through gesture scale data. The gesture recognition here is realized by laser sensor, avoiding the problem of misrecognition caused by visual recognition gestures;
[0019] The recognition of the gesture screen is achieved by using a visual analysis chip to perform matching analysis on the gesture screen data based on a standard gesture feature database stored in a memory;
[0020] The gesture scale data uses a data analysis chip to analyze the trigger data of the laser sensor, thereby efficiently realizing the recognition of gesture corresponding operability and operation value, and is more suitable for the noisy application environment of the machining center;
[0021] Temporary control instructions can be generated efficiently, and before executing the temporary control instructions, the touch display needs to display the operation instructions to be confirmed, and the option of modifying the scale data is provided, thereby achieving more flexible control and avoiding production accidents caused by direct execution due to misidentification.
[0022] The above-mentioned interactive components such as the gesture switch module, motion picture capture module, laser sensor module and touch display should be installed on one side of the transparent window of the machining center. When the operator observes the machining process through the transparent window of the machining center, he can achieve flexible adjustment and control of the machining process through gesture control without leaving the workpiece, thereby providing a more reliable solution for the flexible control of the manufacturing process of the machining center.
[0023] Optionally, the laser sensor module includes a plurality of laser sensors and a liquid crystal digital display arranged in an array, the plurality of laser sensors are arranged at equal intervals, and the liquid crystal digital display is respectively communicatively connected with a numerical control system, a visual analysis chip, a data analysis chip and the plurality of laser sensors; the installation position of the motion picture capture module is located on one side of the laser sensor module.
[0024] By adopting the above technical solution, in actual operation, gesture control is generally for fine-tuning, so the control scale is relatively small, for example, a two-by-two laser sensor array can be used.
[0025] The occlusion of the two laser sensors in the first row is used to identify addition and subtraction, and the occlusion of the two laser sensors in the second row is used to select the number of digits. The LCD digital display can interact with the visual analysis chip and the data analysis chip to operate the corresponding operation items through gestures, and then interact with the numerical control system to see the numerical value corresponding to the current item. The operator can add or subtract the value by blocking the two laser sensors in the first row, and the occlusion of the two laser sensors in the second row is used to select the number of digits, thereby achieving precise gesture modification of the value.
[0026] Optionally, the gesture switch module includes two contact sensors and a single chip microcomputer, and the single chip microcomputer is communicatively connected to the two contact sensors respectively.
[0027] Optionally, the installation distance between the two touch sensors is greater than 100 mm. When the operator's hand swings from the front of any touch sensor to the other touch sensor within a set time, the microcontroller communicates with the motion picture capture module and the laser sensor module to output a signal to start detecting the gesture picture, and the motion picture capture module and the laser sensor module are turned on.
[0028] By adopting the above technical solution, in order to avoid misoperation as much as possible and to prevent the motion picture capture module and the laser sensor module from being in a working state for a long time, a gesture switch module is used to control the switch. Specifically, a control mode of two contact sensors can be adopted. The hand is quickly swept in front of the two contact sensor heads. For example, the microcontroller receives the signals of the two contact sensors within 0.5 seconds, and then communicates with the motion picture capture module and the laser sensor module to output a signal to start detecting the gesture picture.
[0029] Optionally, in step 1, the action picture capture module collects the operator's gesture picture and records the corresponding timestamp when capturing each frame of the picture, and adds the timestamp data to the beginning position of the frame data.
[0030] Optionally, in step 3, during the matching analysis performed by the visual analysis chip, if there is no match greater than the set matching threshold, the standard gesture feature data with the highest matching degree is compared with the gesture screen data in different time periods. If the gesture screen data consisting of a time period with at least two consecutive timestamps has a matching degree greater than the set matching threshold for the time period corresponding to the standard gesture feature data, the match is determined to be successful, and the standard gesture feature data will transmit the corresponding control instruction data to the instruction generation chip.
[0031] By adopting the above technical solution, since the environment of the machining center is relatively harsh, water vapor, smoke and other environments may affect gesture recognition. Here, when the motion picture capture module captures the gesture picture, the timestamp data is synchronously added to the beginning of the frame data;
[0032] When the visual analysis chip performs a matching analysis, if there is no matching degree greater than the set matching degree threshold, the recognition is not directly determined to have failed. Instead, the operator's gesture can be judged based on whether there are gesture screen data consisting of a time period of at least two consecutive timestamps and a matching degree greater than the set matching degree threshold corresponding to the time period of the standard gesture feature data, thereby improving the robustness of gesture recognition.
[0033] Optionally, in step 3, the visual analysis chip uses the following method to perform matching analysis:
[0034] Step a, extracting key features consisting of contours and key points from the captured gesture image;
[0035] Step b, standardizing the extracted key features;
[0036] Step c, matching the standardized key features with a standard gesture feature database stored in a memory;
[0037] Step d: respectively calculating the Euclidean distances between the multiple key feature vectors of the captured gesture features and the multiple standard key feature vectors in the database, and performing matching degree calculation based on the Euclidean distances.
[0038] Optionally, in step d, the formula for calculating the Euclidean distance is:
[0039]
[0040] Where A and B represent the key feature vector of the captured gesture features and the standard multiple key feature vectors in the database, n is the dimension of the feature vector, A i and B i Realize the values of A and B in the i-th dimension respectively;
[0041] The Euclidean distances between all pairs of key feature vectors are calculated and then averaged using the following formula:
[0042]
[0043] Where d(A i , B i ) is the i-th pair of key feature vectors A i and B i , w is the number of key eigenvectors.
[0044] By adopting the above technical solution, the Euclidean distance itself is a distance metric, which directly gives the straight-line distance between two points in space. In matching calculation, the Euclidean distance is usually used as a dissimilarity metric, that is, the smaller the distance, the more similar the two points (or feature vectors) are.
[0045] Optionally, the following formula is used to calculate the matching degree X between the captured gesture picture and the standard gesture picture;
[0046]
[0047] By adopting the above technical solution, to convert the Euclidean distance into a matching degree, since the larger the Euclidean distance is, the less similar it is, we can convert it into a matching degree through a formula. The matching degree X ranges between [0,1], where 1 indicates a complete match and 0 indicates a complete mismatch.
[0048] In summary, the present invention includes at least one of the following beneficial technical effects:
[0049] The present invention can provide a flexible control method for the manufacturing process of a machining center based on motion capture, which adopts a gesture switch module to perform switch control on the entire gesture detection to avoid misoperation. The operator's gesture screen is for interactive operation items, for example, the operation items may be feed speed, program selection control, feed amount adjustment, etc.; gesture scale data can realize direct adjustment of the control value, solving the problem that a single gesture control cannot accurately adjust a specific value, for example, the specific value of the feed amount can be directly adjusted through gesture scale data. The gesture recognition here is realized by laser sensor, avoiding the problem of misrecognition caused by visual recognition gesture; the recognition of gesture screen is realized by matching the gesture screen data with the standard gesture feature database stored in the memory using visual analysis chip; and the gesture scale data uses data analysis chip to analyze the trigger data of laser sensor, so as to efficiently realize the recognition of gesture corresponding operability and operation value, which is more adaptable to the noisy application environment of machining center; temporary control instructions can be generated efficiently, and before executing the temporary control instructions, the operation instructions to be confirmed need to be displayed on the touch display, and the option of modifying the scale data is provided, so as to realize more flexible control and avoid production accidents caused by direct execution due to misrecognition; the operator can realize flexible adjustment control of the machining process through gesture control without leaving the workpiece while observing the machining process through the transparent window of the machining center, which provides a more reliable solution for the flexible control of the manufacturing process of the machining center. BRIEF DESCRIPTION OF THE DRAWINGS
[0050] Figure 1 It is a schematic diagram of the connection principle of electrical components of a flexible control method of a machining center manufacturing process based on motion capture according to the present invention;
[0051] Figure 2 It is a schematic diagram of the connection principle of various electrical components adopted by the flexible control method of the machining center manufacturing process based on motion capture of the present invention.
[0052] Explanation of the accompanying drawings: 1. gesture controller; 11. gesture switch module; 111. contact sensor; 112. single-chip microcomputer; 12. motion picture capture module; 131. visual analysis chip; 132. data analysis chip; 133. memory; 134. instruction generation chip; 14. laser sensor module; 141. laser sensor; 142. liquid crystal digital display; 2. touch display; 100. numerical control system. DETAILED DESCRIPTION
[0053] The present invention is further described in detail below in conjunction with the accompanying drawings.
[0054] The embodiment of the invention discloses a flexible control method of a manufacturing process of a machining center based on motion capture.
[0055] Reference Figure 1 and Figure 2 Embodiment 1, a flexible control method for a machining center manufacturing process based on motion capture, the gesture switch module 11 of the gesture controller 1 detects the front gesture command switch signal, if there is a gesture intervention command signal, the gesture controller 1 adopts the following steps to start recognizing and analyzing the gesture control signal:
[0056] Step 1, the motion picture capture module 12 of the gesture controller 1 first collects the operator's gesture picture, and the laser sensor module 14 then senses the gesture scale data;
[0057] Step 2, the gesture picture data and the gesture scale data are packaged and sent to a gesture instruction analysis module based on the combination of the visual analysis chip 131 and the data analysis chip 132;
[0058] Step 3: The gesture instruction analysis module parses the gesture screen data and the gesture scale data. The visual analysis chip 131 performs a matching degree analysis on the gesture screen data based on the standard gesture feature database stored in the memory 133. If the matching degree is greater than the set matching degree threshold, the corresponding control instruction template data stored in the memory 133 is extracted based on the standard gesture feature, and the corresponding control instruction template data is transmitted to the instruction generation chip 134.
[0059] Step 4, the data analysis chip 132 parses the gesture scale data, analyzes the control scale data contained in the gesture scale data, and transmits the control scale data to the instruction generation chip 134;
[0060] Step 5, the instruction generation chip 134 combines the control scale data with the corresponding control instruction template data to generate a temporary control instruction for controlling the numerical control system 100;
[0061] Step 6, the numerical control system 100 temporarily controls the command to display the text description corresponding to the temporary control command on the gesture command touch display 2, and provides options for modifying the scale data and command confirmation items;
[0062] Step 7, the operator confirms the operation command through the command confirmation item on the gesture command touch display 2, or confirms the temporary control command after fine-tuning through the option of modifying the scale data;
[0063] Step 8: The numerical control system 100 replaces the corresponding part of the existing machining control instruction with the temporary control instruction and executes it.
[0064] The gesture switch module 11 is used to control the entire gesture detection on and off to avoid misoperation. Specifically, it can be a gesture command switch mode, a press switch mode, or an induction switch mode;
[0065] When the gesture switch module 11 detects the front gesture command switch signal, the action picture capture module 12 and the laser sensor module 14, the action picture capture module 12 first collects the operator's gesture picture, and the laser sensor module 14 then senses the gesture scale data;
[0066] The operator's gesture screen is for interactive operation items, such as feed speed, program selection control, feed amount adjustment, etc.; gesture scale data can realize direct adjustment of control values, solving the problem that gesture control alone cannot accurately adjust specific values. For example, the specific value of feed amount can be directly adjusted through gesture scale data. The gesture recognition here is realized by laser sensor, avoiding the problem of misrecognition caused by visual recognition gestures;
[0067] The recognition of the gesture picture is achieved by using the visual analysis chip 131 to perform matching analysis on the gesture picture data based on the standard gesture feature database stored in the memory 133;
[0068] The gesture scale data uses the data analysis chip 132 to analyze the trigger data of the laser sensor, thereby efficiently realizing the recognition of gesture corresponding operability and operation value, and is more suitable for the noisy application environment of the machining center;
[0069] Temporary control instructions can be efficiently generated, and before executing the temporary control instructions, the touch display 2 needs to display the operation instructions to be confirmed, and at the same time, the option of modifying the scale data is provided, thereby achieving more flexible control and avoiding production accidents caused by direct execution due to misidentification.
[0070] The interactive components, including the gesture switch module 11, the motion picture capture module 12, the laser sensor module 14 and the touch display 2, should be installed on one side of the transparent window of the machining center. When the operator observes the machining process through the transparent window of the machining center, the operator can achieve flexible adjustment and control of the machining process through gesture control without leaving the workpiece, thereby providing a more reliable solution for the flexible control of the manufacturing process of the machining center.
[0071] In Example 2, the laser sensor module 14 includes a plurality of laser sensors 141 and a liquid crystal digital display 142 arranged in an array, wherein the plurality of laser sensors 141 are arranged at equal intervals, and the liquid crystal digital display 142 is respectively communicated with the numerical control system 100, the visual analysis chip 131, the data analysis chip 132 and the plurality of laser sensors 141; the installation position of the motion picture capture module 12 is located on one side of the laser sensor module 14.
[0072] In actual operation, gesture control is generally used for fine-tuning, so the scale of control is relatively small. For example, a two-by-two square array of laser sensors 141 can be used. The occlusion of the two laser sensors 141 in the first row is used to identify addition and subtraction, and the occlusion of the two laser sensors 141 in the second row is used to select the number of digits. The LCD digital display 142 can interact with the visual analysis chip 131 and the data analysis chip 132 to interact with the operation items corresponding to the gesture operation, and then interact with the numerical control system 100 to interact with the numerical control system 100 to interact with the numerical control system 100 to interact with the numerical control system 100. The operator adds and subtracts the value by occluding the two laser sensors 141 in the first row, and the occlusion of the two laser sensors 141 in the second row is used to select the number of digits, thereby achieving precise gesture modification of the value.
[0073] In the third embodiment, the gesture switch module 11 includes two contact sensors 111 and a single chip microcomputer 112 . The single chip microcomputer 112 is communicatively connected with the two contact sensors 111 .
[0074] In Example 4, the installation interval between the two touch sensors is greater than 100 mm. When the operator's hand swings from the front of any touch sensor 111 to the other touch sensor 111 within a set time, the microcontroller 112 communicates with the motion picture capture module 12 and the laser sensor module 14 to output a signal to start detecting the gesture picture, and the motion picture capture module 12 and the laser sensor module 14 are turned on.
[0075] In order to avoid misoperation as much as possible and prevent the motion picture capture module 12 and the laser sensor module 14 from being in a working state for a long time, the gesture switch module 11 is used to control the switch. Specifically, it can be a control mode of two contact sensors 111. The hand is quickly swept in front of the sensor heads of the two contact sensors 111. For example, the microcontroller 112 receives the signals of the two contact sensors 111 within 0.5 seconds, and then communicates with the motion picture capture module 12 and the laser sensor module 14 to output a signal to start detecting the gesture picture.
[0076] In Example 5, in step 1, the action picture capturing module 12 collects the operator's gesture picture and records the corresponding timestamp when capturing each frame, and adds the timestamp data to the beginning position of the frame data.
[0077] In Example 6, in step 3, during the matching degree analysis performed by the visual analysis chip 131, if there is no matching degree greater than the set matching degree threshold, the standard gesture feature data with the highest matching degree is taken and compared with the gesture screen data in different time periods. If there are gesture screen data consisting of a time period with at least two consecutive timestamps and a matching degree greater than the set matching degree threshold between the time period corresponding to the standard gesture feature data, the match is determined to be successful, and the standard gesture feature data will transmit the corresponding control instruction data to the instruction generation chip 134.
[0078] Since the environment of the machining center is relatively harsh, water vapor, smoke and other environments may affect gesture recognition. Here, when the motion picture capture module 12 captures the gesture picture, the timestamp data is synchronously added to the beginning of the frame data;
[0079] When the visual analysis chip 131 performs a matching analysis, if there is no matching degree greater than the set matching degree threshold, the recognition is not directly determined to be a failure. Instead, the operator's gesture can be judged based on whether there are gesture screen data consisting of a time period of at least two consecutive timestamps and the matching degree of the time period corresponding to the standard gesture feature data is greater than the set matching degree threshold, thereby improving the robustness of gesture recognition.
[0080] In Example 7, in step 3, the visual analysis chip 131 performs matching degree analysis using the following method:
[0081] Step a, extracting key features consisting of contours and key points from the captured gesture image;
[0082] Step b, standardizing the extracted key features;
[0083] Step c, matching the standardized key features with the standard gesture feature database stored in the memory 133;
[0084] Step d: respectively calculating the Euclidean distances between the multiple key feature vectors of the captured gesture features and the multiple standard key feature vectors in the database, and performing matching degree calculation based on the Euclidean distances.
[0085] In Example 8, in step d, the formula for calculating the Euclidean distance is:
[0086]
[0087] Where A and B represent the key feature vector of the captured gesture features and the standard multiple key feature vectors in the database, n is the dimension of the feature vector, A i and B i Realize the values of A and B in the i-th dimension respectively;
[0088] The Euclidean distances between all pairs of key feature vectors are calculated and then averaged using the following formula:
[0089]
[0090] Where d(A i , B i 0 is the i-th key feature vector A i and B i , w is the number of key eigenvectors.
[0091] The Euclidean distance itself is a distance metric that directly gives the straight-line distance between two points in space. In matching calculations, the Euclidean distance is often used as a dissimilarity metric, that is, the smaller the distance, the more similar the two points (or feature vectors) are.
[0092] In Example 9, the following formula is used to calculate the matching degree X between the captured gesture picture and the standard gesture picture;
[0093]
[0094] To convert the Euclidean distance to a match, since the larger the Euclidean distance, the less similar it is, we can convert it to a match through the formula. The match X ranges between [0,1], where 1 indicates a perfect match and 0 indicates a complete mismatch.
[0095] The above are all preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the protection scope of the present invention.
Claims
1. A flexible control method for manufacturing process of machining center based on motion capture, characterized in that: The gesture switch module (11) of the gesture controller (1) detects the front gesture command switch signal. If there is a gesture intervention command signal, the gesture controller (1) uses the following steps to start identifying and analyzing the gesture control signal: Step 1, the motion picture capture module (12) of the gesture controller (1) first collects the operator's gesture picture, and the laser sensor module (14) then senses the gesture scale data; Step 2, packaging the gesture picture data and the gesture scale data and sending them to a gesture command analysis module based on a combination of a visual analysis chip (131) and a data analysis chip (132); Step 3: The gesture instruction analysis module parses the gesture screen data and the gesture scale data, and the visual analysis chip (131) performs a matching degree analysis on the gesture screen data based on the standard gesture feature database stored in the memory (133). If the matching degree is greater than a set matching degree threshold, the corresponding control instruction template data stored in the memory (133) is extracted based on the standard gesture feature, and the corresponding control instruction template data is transmitted to the instruction generation chip (134); Step 4, the data analysis chip (132) parses the gesture scale data, analyzes the control scale data contained in the gesture scale data, and transmits the control scale data to the instruction generation chip (134); Step 5, the instruction generation chip (134) combines the control scale data with the corresponding control instruction template data to generate a temporary control instruction for controlling the numerical control system (100); Step 6: The temporary control instruction displays a text description corresponding to the temporary control instruction on the gesture instruction touch display (2), and provides an option to modify the scale data and an instruction confirmation item; Step 7, the operator confirms the operation command through the command confirmation item on the gesture command touch display (2), or confirms the temporary control command after fine-tuning through the option of modifying the scale data; Step 8: The numerical control system (100) replaces the corresponding part of the existing machining control instruction with the temporary control instruction and executes it.
2. The flexible control method of a machining center manufacturing process based on motion capture according to claim 1 is characterized in that: The laser sensor module (14) comprises a plurality of laser sensors (141) and a liquid crystal digital display (142) arranged in an array, the plurality of laser sensors (141) being arranged at equal intervals, and the liquid crystal digital display (142) being respectively connected in communication with the numerical control system (100), the visual analysis chip (131), the data analysis chip (132) and the plurality of laser sensors (141); and the installation position of the motion picture capture module (12) is located on one side of the laser sensor module (14).
3. The flexible control method of a machining center manufacturing process based on motion capture according to claim 2 is characterized in that: The gesture switch module (11) comprises two contact sensors (111) and a single-chip microcomputer (112), wherein the single-chip microcomputer (112) is communicatively connected to the two contact sensors (111) respectively.
4. The flexible control method of a machining center manufacturing process based on motion capture according to claim 3 is characterized in that: The installation interval between the two contact sensors is greater than 100 mm. When the operator's hand swings from the front of any contact sensor (111) to the other contact sensor (111) within a set time, the single-chip computer (112) communicates and outputs a signal to start detecting the gesture image to the motion image capture module (12) and the laser sensor module (14), and the motion image capture module (12) and the laser sensor module (14) are turned on.
5. The flexible control method of a machining center manufacturing process based on motion capture according to claim 4 is characterized in that: In step 1, the action picture capturing module (12) collects the operator's gesture picture and records the corresponding timestamp when capturing each frame, and adds the timestamp data to the beginning position of the frame data.
6. The method for flexible control of a machining center manufacturing process based on motion capture according to claim 5 is characterized in that: In step 3, during the matching degree analysis performed by the visual analysis chip (131), if there is no matching degree greater than the set matching degree threshold, the standard gesture feature data with the highest matching degree is taken and compared with the gesture screen data in different time periods. If there are gesture screen data consisting of a time period with at least two consecutive timestamps and a matching degree greater than the set matching degree threshold between the time period corresponding to the standard gesture feature data, the match is determined to be successful, and the control instruction data corresponding to the standard gesture feature data is transmitted to the instruction generation chip (134).
7. The method for flexible control of a machining center manufacturing process based on motion capture according to claim 6 is characterized in that: In step 3, the visual analysis chip (131) performs matching analysis using the following method: Step a, extracting key features consisting of contours and key points from the captured gesture image; Step b, standardizing the extracted key features; Step c, matching the standardized key features with a standard gesture feature database stored in a memory (133); Step d: respectively calculating the Euclidean distances between the multiple key feature vectors of the captured gesture features and the multiple standard key feature vectors in the database, and performing matching degree calculation based on the Euclidean distances.
8. The method for flexible control of a machining center manufacturing process based on motion capture according to claim 7 is characterized in that: In step d, the formula for calculating the Euclidean distance is: Where A and B represent the key feature vector of the captured gesture features and the standard multiple key feature vectors in the database, n is the dimension of the feature vector, A i and B i Realize the values of A and B in the i-th dimension respectively; The Euclidean distances between all pairs of key feature vectors are calculated and then averaged using the following formula: Where d(A i , B i ) is the i-th key feature vector A i and B i , w is the number of key eigenvectors.
9. The method for flexible control of a machining center manufacturing process based on motion capture according to claim 8, characterized in that: The matching degree X between the captured gesture image and the standard gesture image is calculated using the following formula;
Citation Information
Patent Citations
Non-contact type gesture recognition device and Non-contact type gesture recognition method
CN106774887A
Numerical controller
CN107436587A
Visual industrial controller and method based on PLC
CN116430795A
Hand rehabilitation training glove system based on machine vision gesture recognition
CN116807831A
CNC machine tool data acquisition and management system based on big data
CN118192421A
