CN113751429B - High-precision positioning and motion compensation method for mechanical arm assisted laser processing - Google Patents
High-precision positioning and motion compensation method for mechanical arm assisted laser processing Download PDFInfo
- Publication number
- CN113751429B CN113751429B CN202110921612.4A CN202110921612A CN113751429B CN 113751429 B CN113751429 B CN 113751429B CN 202110921612 A CN202110921612 A CN 202110921612A CN 113751429 B CN113751429 B CN 113751429B
- Authority
- CN
- China
- Prior art keywords
- mechanical arm
- layer
- depth
- hidden layer
- output
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B08—CLEANING
- B08B—CLEANING IN GENERAL; PREVENTION OF FOULING IN GENERAL
- B08B7/00—Cleaning by methods not provided for in a single other subclass or a single group in this subclass
- B08B7/0035—Cleaning by methods not provided for in a single other subclass or a single group in this subclass by radiant energy, e.g. UV, laser, light beam or the like
- B08B7/0042—Cleaning by methods not provided for in a single other subclass or a single group in this subclass by radiant energy, e.g. UV, laser, light beam or the like by laser
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B08—CLEANING
- B08B—CLEANING IN GENERAL; PREVENTION OF FOULING IN GENERAL
- B08B13/00—Accessories or details of general applicability for machines or apparatus for cleaning
Landscapes
- Physics & Mathematics (AREA)
- Optics & Photonics (AREA)
- Length Measuring Devices By Optical Means (AREA)
- Numerical Control (AREA)
Abstract
A high-precision positioning and motion compensation method for assisting laser processing of a mechanical arm relates to the field of mechanical arm motion control. The method aims at the problem of large machining errors caused by mismatching of actual removal depth in the composite material with the depth and direction of a machining path planned by a mechanical arm, scans the composite material layer by layer before machining, collects diffuse reflection spectrums of materials at different depths, and establishes a material attribute classification and prediction model related to the depth in the material according to spectral characteristics. The target depth information is compared with diffuse reflection spectrum characteristic information of the target to be removed, which is acquired in the machining process, the depth information of the target to be removed is accurately judged, the three-dimensional coordinate of the actual removal position in the target is accurately calculated by combining the real-time coordinate position in the mechanical arm path planning, and the three-dimensional coordinate is fed back to the mechanical arm machining path for correction in real time through an industrial personal computer, so that accurate positioning and high-precision motion compensation are realized.
Description
Technical Field
The invention relates to the field of mechanical arm motion control, in particular to a high-precision positioning and motion compensation method for auxiliary laser processing of a mechanical arm.
Background
The laser processing is used as a non-contact processing mode, has the characteristics of high precision, small damage and the like, and has unique advantages in removing high-precision and low-damage of the composite material by matching with a mechanical arm controlled by a servo technology. However, the composite material has the characteristics of complex and variable material components and the like, and the absorption rates of materials with different components to laser are different, so that the actual removal depth in the material is not matched with the depth direction of the motion path planned by the mechanical arm and is not adaptive. If the mechanical arm is used for driving the laser to clean and remove the stains with different thicknesses, shapes and components on the surface of the cultural relic, the cultural relic can be easily damaged due to the fact that the actual removal depth is different from the depth direction of the motion path planned by the mechanical arm. Therefore, how to position and monitor the real three-dimensional coordinate of the position of the removed material in real time and perform motion compensation on the mechanical arm becomes a technical bottleneck facing the prior art for improving the precision of the mechanical arm for assisting the laser to remove the composite material.
In patent CN106020024A (published: 2016.10.12), devices such as a vision acquisition device, a laser range finder, a stay cord encoder and the like are adopted, and the three-dimensional coordinate of the tail end of the mechanical arm is calculated by acquiring the stay cord angle, the rope length and the mechanical arm height, so that the motion path of the mechanical arm is compensated and corrected in real time, but the method is only suitable for contact type processing and is not suitable for laser non-contact processing; patent CN111604598A (published: 2020.09.01) combines multi-frame sequence image dynamic transformation and multi-axis drive feeding device space form and position characteristic information to realize accurate alignment between a laser etching focusing plane and a part initial feature point to be processed in a working space, however, the method only solves the problem of positioning and tool setting in the initial stage of laser etching processing, and cannot accurately position the etching position in the processing process; in patent CN111407467A (publication date: 2020.07.14), a spectrometer is used to perform online monitoring on a plasma plume ejected during laser bone processing, so as to optimize processing parameters in real time, however, the tested object is an ejected removal substance, which cannot provide component information of a material to be processed, and thus, a processing position cannot be accurately located, and optimal compensation of a processing path cannot be achieved. Because the focus is in the material in the laser processing process, and the signal at the focus is easily interfered by impurities such as plume and the like generated in the processing process, the signal at the position of the focus is difficult to directly detect or the detected signal is unstable, most of the existing researches and patents are focused on optimizing the process parameters of the laser by directly monitoring the terminal coordinate of the mechanical arm or detecting the spectrum of the plasma plume in the removing process and the like. However, high-precision laser processing cannot be realized by independently optimizing the processing parameters of the laser, the laser processing track more directly influences the region and range of the laser action, and the method for obtaining depth information in the tissue by detecting the diffuse reflection spectrum of the material to be removed has the defects that the corresponding relation between the spectral intensity and the depth needs to be calibrated, the spectral intensity is related to the depth and the attribute of the detected material, and the calibration difficulty is high, so that the method is used for accurately positioning the interior of the composite material in the laser processing process, and the real-time compensation of the mechanical arm processing path is realized, which is not reported at present.
In order to overcome the problems, the invention provides a high-precision positioning and motion compensation method for auxiliary laser processing of a mechanical arm for the first time.
Disclosure of Invention
The invention aims to solve the technical problem of processing errors caused by mismatching of actual removal depth in a composite material with the depth and direction of a motion path planned by a mechanical arm, and provides a high-precision positioning and motion compensation method for the mechanical arm for laser processing. The method is based on the diffuse reflection spectrum detection principle, can position and monitor the real three-dimensional coordinates of the removed material position in real time, compensates the mechanical arm, and has the advantages of high precision and high response speed.
The invention adopts the following scheme to solve the technical problems:
1. a high-precision positioning and motion compensation method for mechanical arm assisted laser processing comprises the following steps:
firstly, scanning the composite material layer by layer before processing, collecting diffuse reflection spectrums of the material at different depths, and preprocessing data;
establishing a classification model related to material properties according to the trend characteristics of the diffuse reflection spectrum, and establishing a prediction model related to depth according to the spectral intensity characteristics;
starting a machining program, driving a laser to perform machining operation by a mechanical arm, simultaneously collecting diffuse reflection spectrum information of a removal position, comparing and analyzing the spectrum information with an established model, acquiring depth information of a target to be removed in the machining process by judging the attribute of the material being removed, and accurately calculating three-dimensional coordinates of the actual removal position in the target by combining with a real-time coordinate position in path planning of the mechanical arm;
and step four, feeding the coordinates obtained by calculation back to the mechanical arm in real time by the industrial personal computer for path correction, thereby realizing precise positioning and compensation.
2. Preferably, in the second step, a classification model related to material properties is established according to the diffuse reflection spectrum trend characteristic data of the materials at different depths, and the general classification model is constructed as follows:
wherein O is k Output results for attribute classification, x i For the corresponding ith spectral trend characteristic data, n is the number of nodes of an input layer and n is equal to the number of the spectral trend characteristic data, l is the number of nodes of an implicit layer and l is log 2 n, m is the number of nodes of the output layer and m is equal to the number of attribute types contained in the known material, k is the dimension of the classification output result, f is the excitation function of the hidden layer
ω ij As weights, ω, of the input layer to the hidden layer jk As weights from hidden layer to output layer, a j For input of layer to hidden layer thresholds, b k The method comprises the steps that a threshold value from a hidden layer to an output layer is adopted, the weight and the threshold value need to be continuously updated in an iterative manner according to error back propagation in a model training process, the iteration frequency is usually 5000-10000 times, and a model with the minimum output error is selected;
and each attribute type O k A corresponding prediction model O of the relative depth of laser ablation needs to be established h The construction of the prediction model is as above:
wherein O is h For the prediction of the relative depth of laser ablation, x p For the corresponding p-th spectral intensity characteristic data, s is the number of nodes of the input layer and is equal to the number of the spectral intensity characteristic data, t is the number of nodes of the hidden layer and t is log 2 s, r is the number of output layer nodes and r is equal to 1, h is the dimension of the prediction result, f is the excitation function of the hidden layer
ω pq For weights of input layer to hidden layer, ω qh For hidden layer to output layer weights, a q For input of layer to hidden layer thresholds, b h A threshold value from a hidden layer to an output layer is adopted, wherein the weight and the threshold value need to be continuously updated iteratively according to error back propagation in the model training process, the iteration times are 1000 times, and a model with the minimum output error is selected; the industrial personal computer needs to classify and output a result O according to attributes during actual processing k Calling a corresponding depth prediction model to perform relative depth O h Predicting, wherein the real-time three-dimensional coordinate position in the mechanical arm path planning is (X, Y, H), and calculating to obtain the accurate three-dimensional coordinate of the actual removal position as (X, Y, O) h + Z), Z being the step of the Z axis of the robot arm, H being the removal depth of the path plan.
3. Preferably, the determination conditions for positioning and compensating the mechanical arm in the step four are as follows:
the relative removal depth calculated by the industrial personal computer is O h The step of the Z axis of the mechanical arm is Z, the required removal depth is H, and the actual removal depth is O h + z when O h When + Z is less than H, the step of the Z axis of the mechanical arm needs to be increased, and when O is less than H h When + Z > H, the step of the Z axis of the mechanical arm needs to be reduced until O h And when + z is equal to H, finishing the processing, thereby realizing real-time positioning and compensation.
4. Preferably, the fourth step of the industrial control machine needs to feed back the coordinates obtained by calculation to the mechanical arm in real time for path correction compensation, and the correction compensation range is +/-2 mm.
Compared with the prior art, the invention has the following advantages:
according to the method, classification models related to material attributes are established according to diffuse reflection spectrum trend characteristic data of materials at different depths, a prediction model of relative laser removal depth related to the material attributes is established according to spectrum intensity characteristics by each classification model, and an industrial personal computer needs to select a corresponding depth prediction model according to attribute classification output results during actual processing to perform depth prediction. The method can accurately calculate the three-dimensional coordinate of the actual removal position of the composite material, and feeds the three-dimensional coordinate back to the mechanical arm for path correction, thereby realizing precise positioning and compensation.
Drawings
Fig. 1 is a flowchart of a high-precision positioning and motion compensation method provided by the present invention.
Detailed Description
The present invention will be described in detail with reference to the following examples. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
The first embodiment is as follows:
the laser cleaning technology is commonly used for removing rust on the metal surface, and a layer of rust with uneven thickness and irregular shape is generated on the metal surface due to long-term external erosion, so that a two-layer composite structure is formed by the rust and the metal. The following takes laser rust removal as a column, and details how to realize high-precision positioning and motion compensation of the laser rust removal mechanical arm are described:
1. firstly, scanning the metal to be derusted layer by layer before processing, namely scanning rust and the metal, acquiring diffuse reflection spectrum data of materials at different depths, and carrying out normalization processing on the data;
2. selecting a wave band with the highest accuracy of a spectrum classification model as input according to the diffuse reflection spectrum trend characteristic data of the rust and the metal, outputting 2 nodes and 10 nodes of a hidden layer, iterating for 5000 times, establishing a classification model related to the organization attribute, inputting 200 nodes according to the spectrum intensity characteristic data, outputting 1 node, 8 nodes of the hidden layer, iterating for 1000 times, and respectively establishing a depth prediction model of the rust and the metal;
3. opening the procedure, drive the laser instrument by the arm and carry out the rust cleaning operation to adopt the mode of blowing the protective gas to blow away the impurity that rust cleaning in-process focus position produced, prevent its interference spectrum signal, set up the spectrum appearance integration time simultaneously and do: 500 mu s, 5 average times and 5 pixel smoothness, synchronously collecting diffuse reflection spectrum information of the removal position, performing normalization processing on the spectrum data, taking full spectrum data as input, performing comparative analysis on the full spectrum data and the model established in the step 2, and selecting a corresponding depth prediction model by judging the attribute of the material being removedType calculation depth information O h And accurately calculating the three-dimensional coordinates (X, Y, O) of the actual removal position in the target material by combining the real-time coordinate positions (X, Y, H) in the path planning of the mechanical arm h + Z), wherein Z is the step of the Z axis of the mechanical arm, and H is the removal depth of the path planning;
4. the industrial personal computer calculates O h And the difference value between + z and H is used for correcting the path of the mechanical arm, so that the precise positioning and compensation of the mechanical arm in the derusting process are realized.
In the embodiment of the present invention, all or part of the steps mentioned above may be performed by driving the relevant hardware by a program, and the program and data may be saved in a hard disk of a computer.
The second embodiment is as follows:
the biological tissue is composed of a multilayer tissue structure, is a typical multilayer composite material, and takes a certain in vitro biological tissue as an example (comprising two layers of a tissue A and a tissue B), and how to realize high-precision positioning and motion compensation of the laser biological tissue removal mechanical arm is explained in detail:
1. firstly, scanning isolated biological tissues in a culture dish layer by layer before processing, namely scanning the tissue A and the tissue B, acquiring diffuse reflection spectrum data of tissues at different depths, and carrying out normalization processing on the data;
2. selecting a wave band with the highest accuracy of a spectral classification model as input according to diffuse reflection spectral trend characteristic data of an organization A and an organization B, carrying out iteration for 10000 times, establishing a classification model related to the organization attribute, inputting 300 nodes according to spectral intensity characteristic data, outputting 1 node, carrying out hidden layer 8 nodes, and carrying out iteration for 1000 times, and respectively establishing depth prediction models corresponding to the organization A and the organization B;
3. starting a processing program, driving a laser to remove by a mechanical arm, and simultaneously setting the integration time of a spectrometer as follows: 800 mu s, the average frequency is 10, the pixel smoothness is 10, the diffuse reflection spectrum information of the removal position is synchronously collected, the spectrum data can be used as input after normalization processing, and is compared and analyzed with the model established in the step 2, and the removal position is judged by judgingOrganizing attributes, selecting corresponding depth prediction model to calculate depth information O h And accurately calculating the three-dimensional coordinates (X, Y, O) of the actual removal position in the target material by combining the real-time coordinate positions (X, Y, H) in the path planning of the mechanical arm h + Z), wherein Z is the step of the Z axis of the mechanical arm, and H is the removal depth of the path planning;
4. the industrial personal computer calculates O h And the difference value between + z and H is used for correcting the path of the mechanical arm, so that the precise positioning and compensation of the mechanical arm in the process of removing the biological tissue by the laser are realized.
In the embodiment of the present invention, all or part of the steps mentioned above may be performed by driving the relevant hardware by a program, and the program and data may be saved in a hard disk of a computer.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (3)
1. A high-precision positioning and motion compensation method for mechanical arm assisted laser processing comprises the following steps:
firstly, scanning the composite material layer by layer before processing, collecting diffuse reflection spectrums of the material at different depths, and preprocessing data;
establishing a classification model related to material properties according to the trend characteristics of the diffuse reflection spectrum, and establishing a prediction model related to depth according to the spectral intensity characteristics;
starting a machining program, driving a laser to perform machining operation by a mechanical arm, simultaneously collecting diffuse reflection spectrum information of a removal position, comparing the spectrum information with the established model for analysis, acquiring depth information of the target to be removed in the machining process by judging the property of the material being removed, and accurately calculating the three-dimensional coordinate of the actual removal position in the target by combining with the real-time coordinate position in the path planning of the mechanical arm;
feeding the coordinates obtained by calculation back to the mechanical arm in real time by the industrial personal computer for path correction, thereby realizing precise positioning and compensation;
the method is characterized in that: step two, establishing a classification model related to material attributes according to diffuse reflection spectrum trend characteristic data of materials at different depths, wherein the general classification model is constructed as follows:
wherein O is k Output results for attribute classification, x i For the corresponding ith spectral trend characteristic data, n is the number of nodes of the input layer and n is equal to the number of the spectral trend characteristic data, l is the number of nodes of the hidden layer and
m is the number of nodes of the output layer and is equal to the number of attribute types contained in the known material, k is the dimension of the classification output result, f is the excitation function of the hidden layer
ω ij As weights, ω, of the input layer to the hidden layer jk As weights from hidden layer to output layer, a j For input of layer to hidden layer thresholds, b k The method comprises the steps that a threshold value from a hidden layer to an output layer is adopted, the weight and the threshold value need to be continuously updated in an iterative manner according to error back propagation in a model training process, the iteration frequency is usually 5000-10000 times, and a model with the minimum output error is selected;
and each attribute type O k A corresponding prediction model O of the relative depth of laser ablation needs to be established h The construction of the prediction model is as above:
wherein O is h For the prediction of the relative depth of laser ablation, x p For the corresponding p-th spectral intensity characteristic data, s is the number of nodes of the input layer and s is equal to the number of the spectral intensity characteristic data, t is the number of nodes of the hidden layer and
r is the number of nodes in the output layer and r is equal to 1, h is the dimension of the prediction result, f is the excitation function of the hidden layer and
ω pq as weights, ω, of the input layer to the hidden layer qh For hidden layer to output layer weights, a q For input of layer to hidden layer thresholds, b h A threshold value from a hidden layer to an output layer is adopted, wherein the weight and the threshold value need to be continuously updated iteratively according to error back propagation in the model training process, the iteration times are 1000 times, and a model with the minimum output error is selected; the industrial personal computer needs to classify and output a result O according to attributes during actual processing k Calling a corresponding depth prediction model to perform relative depth O h Predicting, wherein the real-time three-dimensional coordinate position in the mechanical arm path planning is (X, Y, H), and calculating to obtain the accurate three-dimensional coordinate of the actual removal position as (X, Y, O) h + Z), Z being the step of the Z axis of the robot arm, H being the removal depth of the path plan.
2. The method for high-precision positioning and motion compensation of mechanical arm assisted laser processing according to claim 1, wherein the method comprises the following steps:
step four, the judgment conditions for positioning and compensating the mechanical arm are as follows:
the relative removal depth calculated by the industrial personal computer is O h The step of the Z axis of the mechanical arm is Z, the required removal depth is H, and the actual removal depth is O h + z when O h When + Z is less than H, the step of the Z axis of the mechanical arm needs to be increased, and when O is less than H h When + z > H, it is necessaryReduce the Z-axis step of the robot arm until O h And when + z is equal to H, finishing the processing, thereby realizing real-time positioning and compensation.
3. The method for high-precision positioning and motion compensation of mechanical arm assisted laser processing according to claim 1, wherein the method comprises the following steps:
the industrial personal computer needs to feed back the calculated coordinates to the mechanical arm in real time for path correction compensation, and the correction compensation range is +/-2 mm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110921612.4A CN113751429B (en) | 2021-08-11 | 2021-08-11 | High-precision positioning and motion compensation method for mechanical arm assisted laser processing |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110921612.4A CN113751429B (en) | 2021-08-11 | 2021-08-11 | High-precision positioning and motion compensation method for mechanical arm assisted laser processing |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113751429A CN113751429A (en) | 2021-12-07 |
| CN113751429B true CN113751429B (en) | 2022-08-02 |
Family
ID=78789075
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110921612.4A Active CN113751429B (en) | 2021-08-11 | 2021-08-11 | High-precision positioning and motion compensation method for mechanical arm assisted laser processing |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113751429B (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114850658B (en) * | 2022-05-17 | 2024-05-24 | 北京工业大学 | Laser parameter automatic regulation and control method based on reflection spectrum |
| CN115599107B (en) * | 2022-12-08 | 2023-04-28 | 唐山雄炜机器人有限公司 | Laser automatic rust cleaning robot control system based on artificial intelligence |
| CN116543050B (en) * | 2023-05-26 | 2024-03-26 | 深圳铭创智能装备有限公司 | Transparent curved surface substrate positioning method, computer equipment and storage medium |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5900633A (en) * | 1997-12-15 | 1999-05-04 | On-Line Technologies, Inc | Spectrometric method for analysis of film thickness and composition on a patterned sample |
| WO2010003289A1 (en) * | 2008-07-11 | 2010-01-14 | 中国科学院沈阳自动化研究所 | Apparatus and method for robots tracking appointed path with high precision |
| CN106824923A (en) * | 2017-03-16 | 2017-06-13 | 融之航信息科技（苏州）有限公司 | A kind of composite material surface coated laser cleaning device and its cleaning method |
| CN108672937A (en) * | 2018-04-02 | 2018-10-19 | 西南交通大学 | A kind of real-time monitoring and regulating system of laser-MAG welding penetrations |
| CN108908337A (en) * | 2018-07-26 | 2018-11-30 | 苏州科技大学 | Manipulator repetitive positioning accuracy measuring device and method based on digital speckle interference |
| CN111373221A (en) * | 2017-11-24 | 2020-07-03 | Abb瑞士股份有限公司 | System and method for characterizing coatings, such as paint films, by radiation and painting installation using such a system |
| CN111650187A (en) * | 2020-04-29 | 2020-09-11 | 北京航空航天大学合肥创新研究院 | Laser paint removal real-time feedback and damage inhibition method based on multispectral signal response |
| CN113134837A (en) * | 2021-04-16 | 2021-07-20 | 浙江大学 | Method for improving absolute positioning precision based on six-degree-of-freedom series mechanical arm |
-
2021
- 2021-08-11 CN CN202110921612.4A patent/CN113751429B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5900633A (en) * | 1997-12-15 | 1999-05-04 | On-Line Technologies, Inc | Spectrometric method for analysis of film thickness and composition on a patterned sample |
| WO2010003289A1 (en) * | 2008-07-11 | 2010-01-14 | 中国科学院沈阳自动化研究所 | Apparatus and method for robots tracking appointed path with high precision |
| CN106824923A (en) * | 2017-03-16 | 2017-06-13 | 融之航信息科技（苏州）有限公司 | A kind of composite material surface coated laser cleaning device and its cleaning method |
| CN111373221A (en) * | 2017-11-24 | 2020-07-03 | Abb瑞士股份有限公司 | System and method for characterizing coatings, such as paint films, by radiation and painting installation using such a system |
| CN108672937A (en) * | 2018-04-02 | 2018-10-19 | 西南交通大学 | A kind of real-time monitoring and regulating system of laser-MAG welding penetrations |
| CN108908337A (en) * | 2018-07-26 | 2018-11-30 | 苏州科技大学 | Manipulator repetitive positioning accuracy measuring device and method based on digital speckle interference |
| CN111650187A (en) * | 2020-04-29 | 2020-09-11 | 北京航空航天大学合肥创新研究院 | Laser paint removal real-time feedback and damage inhibition method based on multispectral signal response |
| CN113134837A (en) * | 2021-04-16 | 2021-07-20 | 浙江大学 | Method for improving absolute positioning precision based on six-degree-of-freedom series mechanical arm |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113751429A (en) | 2021-12-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113751429B (en) | High-precision positioning and motion compensation method for mechanical arm assisted laser processing | |
| CN106312397B (en) | A kind of the welding track automatic tracking system and method for laser vision guiding | |
| CN113427168A (en) | Real-time welding seam tracking device and method for welding robot | |
| CN109489580B (en) | On-machine point cloud detection and compensation method for surface processing of aero-engine blade | |
| CN108982546B (en) | Intelligent robot gluing quality detection system and method | |
| CN111062990A (en) | Binocular vision positioning method for underwater robot target grabbing | |
| Haug et al. | Robust laser-stripe sensor for automated weld-seam-tracking in the shipbuilding industry | |
| CN107133565B (en) | Line laser-based laser engraving line feature extraction method | |
| Srinivasu et al. | Minimum zone evaluation of roundness using hybrid global search approach | |
| CN110682159A (en) | Cutter wear state identification method and device | |
| CN115464263A (en) | Automatic tracking method, detection method and device for laser welding seam | |
| CN110328674A (en) | A kind of laser cutting machine people | |
| Pivkin et al. | Optical image inspection of helical surface geometry of drill | |
| CN114353690B (en) | On-line detection device and detection method for roundness of large aluminum alloy annular forging | |
| US20160349729A1 (en) | Method for machining a component on a multi-axis machine tool driven by an nc-controller and apparatus for conducting said method | |
| Zhou et al. | Intelligent guidance programming of welding robot for 3D curved welding seam | |
| CN114025912A (en) | Machining condition search device and machining condition search method | |
| CN116604212A (en) | Robot weld joint identification method and system based on area array structured light | |
| Sun et al. | A vision-based method for dimensional in situ measurement of cooling holes in aero-engines during laser beam drilling process | |
| González et al. | Adaptive edge finishing process on distorted features through robot-assisted computer vision | |
| CN113910001B (en) | Numerical control machine tool space error identification method | |
| CN114066752A (en) | Line-structured light skeleton extraction and burr removal method for weld tracking | |
| Zou et al. | Robotic Seam Tracking System Combining Light-Weight Segmentation Network Design and ADMM-Based Structured Pruning | |
| Leyendecker et al. | Predictive Quality Modeling for Ultra-Short-Pulse Laser Structuring Utilizing Machine Learning | |
| CN116532825A (en) | Laser hole making process monitoring and regulating combined protection method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |