CN112893180A - Object touch classification method and system considering friction coefficient abnormal value elimination - Google Patents
Object touch classification method and system considering friction coefficient abnormal value elimination Download PDFInfo
- Publication number
- CN112893180A CN112893180A CN202110072637.1A CN202110072637A CN112893180A CN 112893180 A CN112893180 A CN 112893180A CN 202110072637 A CN202110072637 A CN 202110072637A CN 112893180 A CN112893180 A CN 112893180A
- Authority
- CN
- China
- Prior art keywords
- friction coefficient
- force
- touch
- obtaining
- sensor
- 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.)
- Pending
Links
- 230000002159 abnormal effect Effects 0.000 title claims abstract description 43
- 238000000034 method Methods 0.000 title claims abstract description 42
- 230000008030 elimination Effects 0.000 title claims abstract description 12
- 238000003379 elimination reaction Methods 0.000 title claims abstract description 12
- 238000004364 calculation method Methods 0.000 claims description 13
- 238000005259 measurement Methods 0.000 claims description 11
- 238000012545 processing Methods 0.000 claims description 4
- 239000000463 material Substances 0.000 abstract description 26
- 230000001133 acceleration Effects 0.000 description 5
- 238000004458 analytical method Methods 0.000 description 5
- 238000004422 calculation algorithm Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 4
- 238000007621 cluster analysis Methods 0.000 description 3
- 230000009897 systematic effect Effects 0.000 description 3
- 238000001514 detection method Methods 0.000 description 2
- 238000011084 recovery Methods 0.000 description 2
- 238000007405 data analysis Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000011985 exploratory data analysis Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/34—Sorting according to other particular properties
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B07—SEPARATING SOLIDS FROM SOLIDS; SORTING
- B07C—POSTAL SORTING; SORTING INDIVIDUAL ARTICLES, OR BULK MATERIAL FIT TO BE SORTED PIECE-MEAL, e.g. BY PICKING
- B07C5/00—Sorting according to a characteristic or feature of the articles or material being sorted, e.g. by control effected by devices which detect or measure such characteristic or feature; Sorting by manually actuated devices, e.g. switches
- B07C5/36—Sorting apparatus characterised by the means used for distribution
- B07C5/361—Processing or control devices therefor, e.g. escort memory
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/21—Design, administration or maintenance of databases
- G06F16/215—Improving data quality; Data cleansing, e.g. de-duplication, removing invalid entries or correcting typographical errors
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/28—Databases characterised by their database models, e.g. relational or object models
- G06F16/284—Relational databases
- G06F16/285—Clustering or classification
Landscapes
- Engineering & Computer Science (AREA)
- Databases & Information Systems (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Quality & Reliability (AREA)
- Testing Of Devices, Machine Parts, Or Other Structures Thereof (AREA)
Abstract
本发明涉及一种考虑摩擦系数异常值剔除的物体触摸分类方法和系统，方法包括：带动触觉传感器在被测物体表面移动，获取法向力和切向力，计算出物体摩擦系数；分别获取各个已知样本的物体摩擦系数，构建物体摩擦系数数据集；对物体摩擦系数数据集进行聚类，获取各聚类组标准值偏差，进行异常值剔除，并建立物体类型与物体摩擦系数的联系数据库；探测待分类物体的多个物体摩擦系数数据，并进行异常值剔除，然后与联系数据库对比，实现物体分类。与现有技术相比，本发明根据不同材质物体的物体摩擦系数不同的特点，实现物体分类，方案方便可靠，效率高；根据聚类组标准值偏差进行了异常值剔除，提高了物体分类结果的准确性。
The invention relates to an object touch classification method and system considering the elimination of abnormal values of friction coefficients. The method includes: driving a tactile sensor to move on the surface of a measured object, obtaining normal force and tangential force, and calculating the object friction coefficient; The object friction coefficient of the sample is known, and the object friction coefficient data set is constructed; the object friction coefficient data set is clustered, the standard value deviation of each cluster group is obtained, the outliers are eliminated, and the connection database between the object type and the object friction coefficient is established. ; Detect the friction coefficient data of multiple objects of the object to be classified, remove outliers, and then compare with the contact database to realize object classification. Compared with the prior art, the present invention realizes object classification according to the characteristics of different friction coefficients of objects of different materials, the scheme is convenient and reliable, and the efficiency is high; the outliers are eliminated according to the deviation of the standard value of the cluster group, and the object classification result is improved. accuracy.
Description
Technical Field
The invention relates to the field of object classification, in particular to an object touch classification method and system considering friction coefficient abnormal value elimination.
Background
With the need of industrial automation process, some methods for recognizing object information and performing classification processing by touching the object through a machine have appeared.
The invention with the publication number of CN108889636A discloses a method for classifying and identifying materials based on the touch of a manipulator, which utilizes the characteristics of collision deformation and deformation recovery of different materials to make the instantaneous acceleration waveform of the manipulator gripping the materials different according to the types of the materials, the acceleration waveform is formed by gripping the materials by the manipulator and collecting the acceleration information when the manipulator grips the materials, the acceleration waveform is matched with the acceleration waveform in a database, so as to obtain the types of the materials gripped by the manipulator at the moment, then the manipulator plans a moving path according to the types of the materials at other parts of consciousness, and the materials are placed in a classification box to finish material classification.
According to the method, materials are classified according to the collision deformation characteristics and the deformation recovery characteristics of different materials, but the surface material of an object or the error of measuring equipment is not considered, so that the accuracy of a classification result cannot be guaranteed.
Disclosure of Invention
The invention aims to provide an object touch classification method and system considering friction coefficient abnormal value rejection in order to overcome the defects that the prior art does not consider the surface material of an object or the error of measuring equipment and cannot ensure the accuracy of a classification result.
The purpose of the invention can be realized by the following technical scheme:
an object touch classification method considering friction coefficient abnormal value elimination comprises the following steps:
an object friction coefficient obtaining step: driving a touch sensor to move on the surface of a measured object, and acquiring touch information through the touch sensor, wherein the touch information comprises a normal force and a tangential force, and the ratio of the normal force to the tangential force is taken as the friction coefficient of the object;
clustering sample data: obtaining a plurality of known samples, obtaining the object friction coefficient of each known sample by respectively adopting the object friction coefficient obtaining step, and constructing an object friction coefficient data set; clustering the object friction coefficient data set, acquiring standard value deviation of each clustering group, removing abnormal values, and establishing a contact database of object types and object friction coefficients;
a touch classification step: obtaining an object to be classified, obtaining a plurality of object friction coefficient data of the object to be classified by adopting the object friction coefficient obtaining step, obtaining standard value deviations of the plurality of object friction coefficient data, removing abnormal values, comparing the rest of object friction coefficient data with the contact database respectively, obtaining an object type corresponding to the object to be classified, and realizing object classification.
Further, the normal force FnThe calculation expression of (a) is:
in the formula, q is the moment of a contact point measured by the touch sensor, and n is the normal vector of the contact surface of the touch sensor;
the tangential force FtThe calculation expression of (a) is:
Ft＝p-Fn
wherein p is the three-dimensional force measured by the tactile sensor.
Further, the normal force and the tangential force are solved by adopting a nonlinear least square method.
Further, the tactile sensor is a six-axis force-torque sensor.
Further, the SPSS system is used for calculating the standard value deviation of the data, so that the abnormal value elimination is carried out.
The invention also provides an object touch classification system considering friction coefficient abnormal value rejection, which comprises a touch sensor, a mechanical arm and a controller, wherein the touch sensor and the mechanical arm are both connected with the controller, the mechanical arm is connected with the touch sensor, and the data processing process of the controller comprises the following steps:
an object friction coefficient obtaining step: driving a touch sensor to move on the surface of a measured object, and acquiring touch information through the touch sensor, wherein the touch information comprises a normal force and a tangential force, and the ratio of the normal force to the tangential force is taken as the friction coefficient of the object;
clustering sample data: obtaining a plurality of known samples, obtaining the object friction coefficient of each known sample by respectively adopting the object friction coefficient obtaining step, and constructing an object friction coefficient data set; clustering the object friction coefficient data set, acquiring standard value deviation of each clustering group, removing abnormal values, and establishing a contact database of object types and object friction coefficients;
a touch classification step: obtaining an object to be classified, obtaining a plurality of object friction coefficient data of the object to be classified by adopting the object friction coefficient obtaining step, obtaining standard value deviations of the plurality of object friction coefficient data, removing abnormal values, comparing the rest of object friction coefficient data with the contact database respectively, obtaining an object type corresponding to the object to be classified, and realizing object classification.
Further, the tactile sensor includes a hemispherical sensor housing and a six-axis force-torque sensor coupled to each other, the six-axis force-torque sensor measuring force-torque measurements generated by the hemispherical sensor housing, the controller calculating coordinates, normal force, and tangential force of a contact point on the hemispherical sensor housing based on the force-torque measurements.
Further, the normal force FnThe calculation expression of (a) is:
in the formula, q is the moment of a contact point measured by the touch sensor, and n is the normal vector of the contact surface of the touch sensor;
the tangential force FtThe calculation expression of (a) is:
Ft＝p-Fn
wherein p is the three-dimensional force measured by the tactile sensor.
Further, the normal force and the tangential force are solved by adopting a nonlinear least square method.
Further, the SPSS system is used for calculating the standard value deviation of the data, so that the abnormal value elimination is carried out.
Compared with the prior art, the invention has the following advantages:
(1) the method and the device have the advantages that the tactile sensor is driven to move on the surface of the object to be tested, the tactile information comprising the normal force and the tangential force is obtained, so that the friction coefficient of the object is obtained, the object classification is realized through cluster analysis according to the characteristics of different object friction coefficients of objects made of different materials, the scheme is convenient and reliable, and the efficiency is high; and (3) as for sample data of the friction coefficient of the object, eliminating abnormal values by calculating the standard value deviation of the clustering group, eliminating the influence of the surface material of the object or the error of measuring equipment, and improving the accuracy of the object classification result.
(2) An anthropomorphic touch method and a touch sensing part are designed, and the touch sensor is used for anthropomorphic touch to acquire a large amount of data such as contact normal force, tangential force and the like suitable for analysis, so that the method is used for object classification, and is simple and convenient in operation process and strong in stability.
(3) The SPSS system is used for clustering the standard value deviation of the group to mark and remove abnormal values, so that the deviation of measured data and friction coefficient calculation results caused by mechanical and object surface reasons can be reduced.
(4) Analyzing contact surfaces of different materials by using an SPSS system clustering algorithm, and establishing a relation between the different materials and corresponding friction coefficients; the object to be detected is identified through cluster analysis of the touch information, and the meaning of touch detection is further developed by identifying the function of the material of the current contact surface, namely the roughness of the contact surface can be sensed through touch, so that the scheme is convenient and reliable, and the efficiency is high.
Drawings
FIG. 1 is a schematic diagram of the overall structure of an object touch classification system with consideration of elimination of abnormal friction coefficient values in the embodiment of the present invention;
FIG. 2 is a schematic structural diagram of a tactile sensor in an embodiment of the invention;
FIG. 3 is a tree diagram illustrating the use of average joins in an embodiment of the present invention;
in the figure, the sensor comprises a sensor shell 1, a six-axis force-torque sensor 2, a mechanical arm 3, a hemispherical sensor shell 4, a sensor strain gauge 5, a sensor base 6 and a sensor base.
Detailed Description
The invention is described in detail below with reference to the figures and specific embodiments. The present embodiment is implemented on the premise of the technical solution of the present invention, and a detailed implementation manner and a specific operation process are given, but the scope of the present invention is not limited to the following embodiments.
Example 1
The embodiment provides an object touch classification method considering friction coefficient abnormal value elimination, which comprises the following steps:
an object friction coefficient obtaining step: the method comprises the steps that a touch sensor is driven to move on the surface of a measured object, touch information is obtained through the touch sensor, the touch information comprises a normal force and a tangential force, and the ratio of the normal force to the tangential force is used as the friction coefficient of the object;
clustering sample data: obtaining a plurality of known samples, obtaining the object friction coefficient of each known sample by adopting an object friction coefficient obtaining step respectively, and constructing an object friction coefficient data set; clustering the object friction coefficient data set, acquiring standard value deviation of each clustering group, removing abnormal values, and establishing a contact database of object types and object friction coefficients;
a touch classification step: the method comprises the steps of obtaining an object to be classified, obtaining a plurality of object friction coefficient data of the object to be classified by adopting an object friction coefficient obtaining step, obtaining standard value deviations of the plurality of object friction coefficient data, removing abnormal values, comparing the rest of object friction coefficient data with a contact database respectively, obtaining an object type corresponding to the object to be classified, and realizing object classification.
Normal force FnThe calculation expression of (a) is:
in the formula, q is the moment of a contact point measured by the touch sensor, and n is the normal vector of the contact surface of the touch sensor;
tangential force FtThe calculation expression of (a) is:
Ft＝p-Fn
wherein p is the three-dimensional force measured by the tactile sensor.
As a preferred embodiment, a non-linear least squares solution is used for the normal and tangential forces.
In a preferred embodiment, the tactile sensor is a six-axis force-torque sensor.
As a preferred embodiment, the outlier rejection is performed by calculating standard value deviations of data using SPSS system (Statistical Product and Service Solutions software).
The embodiment also provides an object touch classification system considering friction coefficient abnormal value rejection, which comprises a touch sensor, a mechanical arm and a controller, wherein the touch sensor and the mechanical arm are both connected with the controller, the mechanical arm is connected with the touch sensor, and the data processing process of the controller comprises the following steps:
an object friction coefficient obtaining step: the method comprises the steps that a touch sensor is driven to move on the surface of a measured object, touch information is obtained through the touch sensor, the touch information comprises a normal force and a tangential force, and the ratio of the normal force to the tangential force is used as the friction coefficient of the object;
clustering sample data: obtaining a plurality of known samples, obtaining the object friction coefficient of each known sample by adopting an object friction coefficient obtaining step respectively, and constructing an object friction coefficient data set; clustering the object friction coefficient data set, acquiring standard value deviation of each clustering group, removing abnormal values, and establishing a contact database of object types and object friction coefficients;
a touch classification step: the method comprises the steps of obtaining an object to be classified, obtaining a plurality of object friction coefficient data of the object to be classified by adopting an object friction coefficient obtaining step, obtaining standard value deviations of the plurality of object friction coefficient data, removing abnormal values, comparing the rest of object friction coefficient data with a contact database respectively, obtaining an object type corresponding to the object to be classified, and realizing object classification.
As a preferred embodiment, the tactile sensor includes a hemispherical sensor housing and a six-axis force-moment sensor coupled to each other, the six-axis force-moment sensor for measuring a force-moment measurement generated by the hemispherical sensor housing, and the controller calculates coordinates of the contact point on the hemispherical sensor housing, a normal force, and a tangential force based on the force-moment measurement.
As a preferred embodiment, the normal force FnThe calculation expression of (a) is:
in the formula, q is the moment of a contact point measured by the touch sensor, and n is the normal vector of the contact surface of the touch sensor;
tangential force FtThe calculation expression of (a) is:
Ft＝p-Fn
wherein p is the three-dimensional force measured by the tactile sensor.
As a preferred embodiment, a non-linear least squares solution is used for the normal and tangential forces.
As a preferred embodiment, the outlier rejection is performed by calculating a standard value deviation of the data using the SPSS system.
A combination of the above preferred embodiments can provide a preferred embodiment, which is described in detail below.
The embodiment provides an object touch classification method considering friction coefficient abnormal value rejection, which is used for realizing the classification of the material of a contact surface according to the friction coefficient through anthropomorphic touch, the abnormal value rejection of the roughness of the contact surface and the cluster analysis of data based on the contact surface characteristics of different materials.
Summary of the invention
The method designs an anthropomorphic touch flow through the touch sensor combined with the mechanical arm, tests different contact surface materials on the basis, and obtains data. By utilizing the characteristics that different contact surface materials are different and the relation between the normal direction and the tangential force of a touch device at a contact position during touch is different, the ratio of the normal direction measured value and the tangential force measured value acquired by touch is preprocessed by using SPSS (V21), abnormal values with overlarge data and overlarge data due to the surface reasons of the device and an object are eliminated, and then the abnormal values are evaluated and classified after system clustering analysis is carried out.
Second, detailed description
2.1, product side
The application mode of the product is as follows: the device combining the mechanical arm Dobot magic and the six-axis force-moment touch sensor ATI nano 17force/torque sensor is designed in the aspect of anthropomorphic touch data acquisition. The mechanical arm drives the sensor to contact with the surface to be detected, the touch information returned by the touch sensor is input into the upper computer, and the upper computer controls the mechanical arm to drive the sensor to move on the contact surface to acquire the touch information. And preprocessing abnormal values by using the ratio of the normal force measurement value to the tangential force measurement value acquired by touch through SPSS (V21), removing the abnormal values with overlarge and overlarge data caused by the surface reasons of equipment and objects, and then carrying out system clustering analysis and evaluation classification.
The data acquisition process by anthropomorphic touch comprises the following steps: when the sensing part contacts with the outside along with the motion of the mechanical arm, the six-axis force-torque sensor sends a force-torque measurement value to the upper computer, the upper computer calculates data through a calculation algorithm to obtain the coordinate of a contact point on the shell, and meanwhile, important information such as the normal force and the tangential force of the shell and an object on the contact position is obtained. The mechanical arm is based on the calculated haptic information, including the normal pressure and tangential force at the point of contact.
The overall structure of the tactile sensing section and the structure of the sensor section are shown in fig. 1 and 2. The integral structure of the touch sensing part comprises a touch sensor, a mechanical arm and a controller, wherein the touch sensor and the mechanical arm are both connected with the controller, and the mechanical arm is connected with the touch sensor.
The tactile sensor includes a hemispherical sensor housing and a six-axis force-torque sensor coupled to each other for measuring force-torque measurements produced by the hemispherical sensor housing, the controller calculating coordinates, normal force and tangential force of a contact point on the hemispherical sensor housing based on the force-torque measurements. Specifically, a six-axis force-torque sensor includes a sensor strain gage and a sensor mount interconnected.
2.2, technical side
In the method for classifying object touches in consideration of friction coefficient abnormal value rejection in the embodiment, the functions of each component include:
a sensor housing: the sensor can also sense the stress from the shell because the shell is fixed on the sensor by the screw;
six-axis force-torque sensor: when the shell generates micro deformation, the strain gauge on the sensor is pressed, and the stress is converted into a difference signal;
mechanical arm: and (3) solving data obtained by the six-axis force-moment sensor by using upper computer software and a nonlinear least square method to obtain information such as normal force, tangential force and the like on the contact position.
Under a static friction model, the following tactile information is deduced by utilizing the stress relation between the shell and the sensor to solve a nonlinear equation set, then a normal vector n is solved by a curved surface equation, and a normal force F is solved by the component of the acting force of the contact point in the normal direction and the relation solved by the component and the moment q of the contact pointn：
In the formula, q is the moment of a contact point measured by the touch sensor, and n is the normal vector of the contact surface of the touch sensor;
and subtracting the three-dimensional acting force p vector measured by the strain gauge to obtain a tangential force:
Ft＝p-Fn
wherein p is the three-dimensional force measured by the tactile sensor.
Calculating and classifying friction factors of different contact surface materials:
when the sensing portion moves across the contact surface, return data for tangential and normal forces are obtained according to the anthropomorphic touch method. A large amount of data are obtained as samples on the basis of carrying out multiple times of anthropomorphic touch on contact surfaces made of different materials, the relation between data obtained by analyzing different contact surfaces by using SPSS (V21) to the ratio of tangential force to normal force in the data, namely the friction coefficient, is searched through the standard value deviation of a cluster group among the data, and the effect of eliminating abnormal numerical values is achieved. The process is designed to quickly detect abnormal cases for data auditing and prioritize any inferential data analysis in the exploratory data analysis step. This algorithm is designed as "anomaly detection"; i.e., the definition of an exception case is not specified as any particular application. And outputting an abnormal reason list which is used for displaying the individual case number, the reason variable, the variable influence value, the variable value and the standard value of the variable of each reason, and then rejecting the abnormal individual case. The data of the causes of abnormal cases in this embodiment are shown in table 1.
TABLE 1 abnormal case reason List
The reason is as follows: 1
And after the elimination, the SPSS (V21) is used for systematic clustering classification. This process can identify relatively uniform groups of cases based on selected characteristics using an algorithm that combines clusters starting with each case in a separate cluster until a class remains. The method of systematic clustering selects a systematic clustering analysis metric for distance data, the squared euclidean distance, i.e., the sum of squared differences between terms, and this measure of non-similarity can be used for distance data. A tree diagram is displayed, as shown in fig. 3, which can be used to evaluate the cohesion of the formed clusters and can provide information about the appropriate number of clusters to retain.
The foregoing detailed description of the preferred embodiments of the invention has been presented. It should be understood that numerous modifications and variations could be devised by those skilled in the art in light of the present teachings without departing from the inventive concepts. Therefore, the technical solutions available to those skilled in the art through logic analysis, reasoning and limited experiments based on the prior art according to the concept of the present invention should be within the scope of protection defined by the claims.
Claims (10)
1. An object touch classification method considering friction coefficient abnormal value elimination is characterized by comprising the following steps:
an object friction coefficient obtaining step: driving a touch sensor to move on the surface of a measured object, and acquiring touch information through the touch sensor, wherein the touch information comprises a normal force and a tangential force, and the ratio of the normal force to the tangential force is taken as the friction coefficient of the object;
said normal force FnThe calculation expression of (a) is:
in the formula, q is the moment of a contact point measured by the touch sensor, and n is the normal vector of the contact surface of the touch sensor;
the tangential force FtThe calculation expression of (a) is:
Ft＝p-Fn
in the formula, p is a three-dimensional acting force measured by the touch sensor;
clustering sample data: obtaining a plurality of known samples, obtaining the object friction coefficient of each known sample by respectively adopting the object friction coefficient obtaining step, and constructing an object friction coefficient data set; clustering the object friction coefficient data set, acquiring standard value deviation of each clustering group, removing abnormal values, and establishing a contact database of object types and object friction coefficients;
a touch classification step: obtaining an object to be classified, obtaining a plurality of object friction coefficient data of the object to be classified by adopting the object friction coefficient obtaining step, obtaining standard value deviations of the plurality of object friction coefficient data, removing abnormal values, comparing the rest of object friction coefficient data with the contact database respectively, obtaining an object type corresponding to the object to be classified, and realizing object classification.
2. The method for classifying object touches considering friction coefficient outlier rejection of claim 1, wherein said normal and tangential forces are solved using a non-linear least squares method.
3. The method for classifying object touches considering friction coefficient outlier rejection of claim 1, wherein said touch sensor is a six-axis force-moment sensor.
4. The method for classifying object touches considering friction coefficient outlier rejection of claim 1, wherein a mechanical arm drives a touch sensor to move on the surface of the object to be measured.
5. The method for classifying object touches considering friction coefficient outlier rejection of claim 1, wherein said outlier rejection is performed by calculating a standard value deviation of data using an SPSS system.
6. An object touch classification system considering friction coefficient abnormal value elimination is characterized by comprising a touch sensor, a mechanical arm and a controller, wherein the touch sensor and the mechanical arm are both connected with the controller, the mechanical arm is connected with the touch sensor, and the data processing process of the controller comprises the following steps:
an object friction coefficient obtaining step: driving a touch sensor to move on the surface of a measured object, and acquiring touch information through the touch sensor, wherein the touch information comprises a normal force and a tangential force, and the ratio of the normal force to the tangential force is taken as the friction coefficient of the object;
clustering sample data: obtaining a plurality of known samples, obtaining the object friction coefficient of each known sample by respectively adopting the object friction coefficient obtaining step, and constructing an object friction coefficient data set; clustering the object friction coefficient data set, acquiring standard value deviation of each clustering group, removing abnormal values, and establishing a contact database of object types and object friction coefficients;
a touch classification step: obtaining an object to be classified, obtaining a plurality of object friction coefficient data of the object to be classified by adopting the object friction coefficient obtaining step, obtaining standard value deviations of the plurality of object friction coefficient data, removing abnormal values, comparing the rest of object friction coefficient data with the contact database respectively, obtaining an object type corresponding to the object to be classified, and realizing object classification.
7. The system of claim 6, wherein the touch sensor comprises a hemispherical sensor housing and a six-axis force-moment sensor connected to each other, the six-axis force-moment sensor is configured to measure a force-moment measurement generated by the hemispherical sensor housing, and the controller is configured to calculate coordinates, normal force and tangential force of a contact point on the hemispherical sensor housing based on the force-moment measurement.
8. The system of claim 7, wherein the normal force F is an abnormal value of friction coefficient and the abnormal value of friction coefficient is eliminatednThe calculation expression of (a) is:
in the formula, q is the moment of a contact point measured by the touch sensor, and n is the normal vector of the contact surface of the touch sensor;
the tangential force FtThe calculation expression of (a) is:
Ft＝p-Fn
wherein p is the three-dimensional force measured by the tactile sensor.
9. The system of claim 8, wherein the normal force and the tangential force are solved by a non-linear least square method.
10. The system for touch classification of an object considering friction coefficient outlier rejection of claim 7, wherein said outlier rejection is performed by calculating a standard value deviation of data using an SPSS system.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110072637.1A CN112893180A (en) | 2021-01-20 | 2021-01-20 | Object touch classification method and system considering friction coefficient abnormal value elimination |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110072637.1A CN112893180A (en) | 2021-01-20 | 2021-01-20 | Object touch classification method and system considering friction coefficient abnormal value elimination |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112893180A true CN112893180A (en) | 2021-06-04 |
Family
ID=76116226
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110072637.1A Pending CN112893180A (en) | 2021-01-20 | 2021-01-20 | Object touch classification method and system considering friction coefficient abnormal value elimination |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112893180A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114394469A (en) * | 2022-01-20 | 2022-04-26 | 德生纺织印染(安庆)有限公司 | Intelligent cloth front and back recognition system |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| AU2316888A (en) * | 1987-09-17 | 1989-04-17 | Cyprus Industrial Minerals Company | Method and apparatus for friction sorting of particulate materials |
| CN107463952A (en) * | 2017-07-21 | 2017-12-12 | 清华大学 | A kind of object material sorting technique based on multi-modal fusion deep learning |
| CN108889636A (en) * | 2018-07-20 | 2018-11-27 | 华侨大学 | A method of based on mechanical hand feeling Classification and Identification material |
| CN110889442A (en) * | 2019-11-20 | 2020-03-17 | 北京工业大学 | A method for object material classification for pulsed ToF depth cameras |
| CN111590611A (en) * | 2020-05-25 | 2020-08-28 | 北京具身智能科技有限公司 | Article classification and recovery method based on multi-mode active perception |
| CN112197676A (en) * | 2020-10-14 | 2021-01-08 | 同济大学 | Method, device, computer equipment and storage medium for acquiring object surface information |
-
2021
- 2021-01-20 CN CN202110072637.1A patent/CN112893180A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| AU2316888A (en) * | 1987-09-17 | 1989-04-17 | Cyprus Industrial Minerals Company | Method and apparatus for friction sorting of particulate materials |
| CN107463952A (en) * | 2017-07-21 | 2017-12-12 | 清华大学 | A kind of object material sorting technique based on multi-modal fusion deep learning |
| CN108889636A (en) * | 2018-07-20 | 2018-11-27 | 华侨大学 | A method of based on mechanical hand feeling Classification and Identification material |
| CN110889442A (en) * | 2019-11-20 | 2020-03-17 | 北京工业大学 | A method for object material classification for pulsed ToF depth cameras |
| CN111590611A (en) * | 2020-05-25 | 2020-08-28 | 北京具身智能科技有限公司 | Article classification and recovery method based on multi-mode active perception |
| CN112197676A (en) * | 2020-10-14 | 2021-01-08 | 同济大学 | Method, device, computer equipment and storage medium for acquiring object surface information |
Non-Patent Citations (2)
| Title |
|---|
| 方静: "基于LRF_ELM算法的研究及其在物体材质分类中的应用", 《中国优秀博硕士学位论文全文数据库-信息科技辑》 * |
| 柴明锐: "《数据挖掘技术及在石油地质中的应用》", 30 September 2017 * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114394469A (en) * | 2022-01-20 | 2022-04-26 | 德生纺织印染(安庆)有限公司 | Intelligent cloth front and back recognition system |
| CN114394469B (en) * | 2022-01-20 | 2023-10-20 | 德生纺织印染(安庆)有限公司 | Cloth intelligent recognition front and back face system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Song et al. | A novel texture sensor for fabric texture measurement and classification | |
| Liu et al. | Surface material recognition through haptic exploration using an intelligent contact sensing finger | |
| WO2018103130A1 (en) | Online quality inspection method for molded products | |
| CN101907448B (en) | Depth measurement method based on binocular three-dimensional vision | |
| Kolamuri et al. | Improving grasp stability with rotation measurement from tactile sensing | |
| CN108572880B (en) | Abnormality diagnosis system for equipment | |
| Sui et al. | Incipient slip detection method with vision-based tactile sensor based on distribution force and deformation | |
| Falco et al. | Performance metrics and test methods for robotic hands | |
| JP2022020722A5 (en) | ||
| Zhang et al. | A bearing fault diagnosis method based on multiscale dispersion entropy and GG clustering | |
| CN109115877A (en) | A kind of camber mode damnification recognition method based on DIC technology | |
| TWM604905U (en) | Intelligent monitoring system applied to stamping process | |
| Markert et al. | Fingertip 6-axis force/torque sensing for texture recognition in robotic manipulation | |
| Palermo et al. | Automatic fracture characterization using tactile and proximity optical sensing | |
| CN112893180A (en) | Object touch classification method and system considering friction coefficient abnormal value elimination | |
| CN105045220B (en) | A kind of method of quality control based on multivariable Z score quality control chart for being used for laboratory diagnosis field or field of industrial production | |
| CN116652971A (en) | Method for identifying objects with different rigidities and controlling grabbing force of manipulator | |
| CN109341838B (en) | Double-robot cooperation system and method for weighing apparatus verification | |
| Kurita et al. | A fingerprint pointing device utilizing the deformation of the fingertip during the incipient slip | |
| CN108132256A (en) | A kind of welding quality machine vision high-precision detecting method | |
| CN117515131A (en) | Method, device, storage medium and equipment for planetary reducer wear monitoring | |
| CN103020598B (en) | A kind ofly carry out knowledge method for distinguishing for attitude variation when people being completed to specific action | |
| Zhao et al. | Center of mass and friction coefficient exploration of unknown object for a robotic grasping manipulation | |
| CN112802182B (en) | Method and system for reconstructing anthropomorphic touch object based on touch sensor | |
| CN209148136U (en) | A Dual Robot Collaborative System for Weighing Instrument Verification |
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 | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20210604 |
|
| RJ01 | Rejection of invention patent application after publication |