CN113043313A - Parallel two-finger mechanical gripper and method for identifying type of gripped object by same - Google Patents
Parallel two-finger mechanical gripper and method for identifying type of gripped object by same Download PDFInfo
- Publication number
- CN113043313A CN113043313A CN202110290016.0A CN202110290016A CN113043313A CN 113043313 A CN113043313 A CN 113043313A CN 202110290016 A CN202110290016 A CN 202110290016A CN 113043313 A CN113043313 A CN 113043313A
- Authority
- CN
- China
- Prior art keywords
- elastic
- parallel
- matrix
- sample
- force 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
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J15/00—Gripping heads and other end effectors
- B25J15/08—Gripping heads and other end effectors having finger members
Landscapes
- Engineering & Computer Science (AREA)
- Robotics (AREA)
- Mechanical Engineering (AREA)
- Manipulator (AREA)
Abstract
The invention discloses a parallel two-finger mechanical claw which comprises a driver and two parallel clamping claws, wherein the two clamping claws are connected to the driver, the driver is used for driving at least one clamping claw to move, a plurality of elastic protruding parts are respectively arranged on the oppositely arranged inner sides of the two clamping claws in an array mode, and cavities are formed in the elastic protruding parts so that the elastic protruding parts stressed when the two clamping claws clamp an object can be retracted inwards. The invention also discloses a method for identifying the type of the grabbed object by using the parallel two-finger mechanical gripper. The invention has good stability and adaptability when grabbing objects with different shapes.
Description
Technical Field
The invention relates to the technical field of mechanical arms, in particular to a parallel two-finger mechanical claw and a method for identifying the type of a grabbed object.
Background
With the rapid development of scientific technology, robots have gradually moved from traditional industrial applications to new areas of "collaboration with humans". Since the environment of the ordinary people is often an unstructured environment, the perception capability and the flexibility of the robot are greatly challenged. An actuating mechanism (manipulator) is an important guarantee for the robot to realize various complex operations. Therefore, the design of the novel mechanical claw has great significance for the development of the robot technology.
The current mechanical claw is divided into a gear clamping jaw, a parallel clamping jaw, a bionic multi-finger clamping jaw and other structures according to the structure. Parallel structure's gripper compares and indicates the structure bigger area of contact more than many, but current parallel structure's arm only is fit for snatching the shape rule, not fragile, non-deformable's object, snatchs to some anomalous objects, leads to the object atress inhomogeneous easily to make the object landing, snatch fragile object and be the size of the control power that can't be fine more.
The above background disclosure is only for the purpose of assisting understanding of the concept and technical solution of the present invention and does not necessarily belong to the prior art of the present patent application, and should not be used for evaluating the novelty and inventive step of the present application in the case that there is no clear evidence that the above content is disclosed at the filing date of the present patent application.
Disclosure of Invention
In order to solve the technical problems, the invention provides the parallel two-finger mechanical gripper and the method for identifying the type of the gripped object, and the parallel two-finger mechanical gripper has good stability and adaptability when gripping objects with different shapes.
In order to achieve the purpose, the invention adopts the following technical scheme:
the invention discloses a parallel two-finger mechanical claw, which comprises a driver and two parallel clamping claws, wherein the two clamping claws are connected to the driver, the driver is used for driving at least one clamping claw to move, a plurality of elastic protruding parts are respectively arranged on the oppositely arranged inner sides of the two clamping claws in an array manner, and cavities are arranged in the elastic protruding parts so that the elastic protruding parts stressed when the two clamping claws clamp an object can be retracted inwards.
Preferably, each of the elastic protrusions has an outer diameter that is gradually larger at least in part in a direction from the tip portion to the root portion.
Preferably, each elastic protrusion is of a layered telescopic structure, and the inner diameter of each layer of telescopic structure is larger and larger from the top to the root of the elastic protrusion.
Preferably, the cavities of the elastic protrusions arranged on each clamping jaw are communicated with at least one air cavity, and the outlet of each air cavity is respectively connected with an air pipe so as to inflate and deflate the air cavity through the air pipe. The cavity of the elastic protrusion part is communicated with the air cavity, so that the elastic protrusion part array capable of controlling the size of elasticity through air pressure is realized, the air pressure can be changed according to the shape and the hardness degree of an object to control the size of the elasticity of the elastic protrusion part, and the grabbing of different objects is realized.
Preferably, at least one group of air cavity partition plates are arranged in each clamping jaw to divide the cavity in the clamping jaw into a plurality of air cavities, and the cavities of the elastic protrusions are respectively communicated with the air cavities corresponding to the positions of the elastic protrusions.
Preferably, each clamping jaw is further provided with a force sensor array, and the position of the force sensor array corresponds to the position of the plurality of elastic protrusions arranged in an array; and an elastic cylinder extending to the root of each elastic protrusion is arranged on the inner wall of the top of each elastic protrusion, so that the end part of the elastic cylinder corresponding to the elastic protrusion stressed when the two clamping jaws clamp the object can be in contact with the force sensor array. The method of combining the pneumatic elastic protruding part and the force sensor array is adopted, and a layer of elastic protruding part capable of controlling the elastic force by using the air pressure is arranged on the surface of the force sensor, so that the pressure detection is realized; in addition, the mechanical claw can detect the shape of the object by sensing the position where the elastic protrusion is compressed and the magnitude of the force.
Preferably, the length of the elastic cylinder is smaller than the length of the elastic protrusion.
Preferably, the force sensor array is a piezoresistive force sensor array.
The invention also discloses a method for identifying the type of the grabbed object by adopting the parallel two-finger mechanical gripper, which comprises the following steps:
s1: providing N types of objects, grabbing each type of object M times by the two parallel fingers, and respectively collecting a matrix of forces borne by the force sensor arrays of the two clamping jaws to form a sample matrix when grabbing the object by the two parallel fingers each time to obtain N x M sample matrices;
s2: calculating Euclidean distance Dist (x, y) of any two sample matrixes:
wherein x and y are any two of the sample matrixes respectively, xi、yiRespectively representing the ith element of the x and y matrixes, and n is the sum of the number of the elastic protrusions of the two clamping jaws;
s3: sequencing each Euclidean distance, and dividing samples corresponding to M sample matrixes with the minimum Euclidean distance into one class and N classes in total;
s4: classifying the M samples in each class into the class with the highest proportion of the M samples to form a KNN classifier;
s5: the method comprises the steps of grabbing an object to be classified by the parallel two-finger mechanical gripper, collecting a matrix of forces borne by the force sensor arrays of the two clamping jaws to form a sample matrix to be classified, putting the sample matrix to be classified into the KNN classifier, and selecting the class closest to the Euclidean distance of the sample matrix to be classified as the class of the object to be classified by the KNN classifier.
Preferably, before step S2 is executed, N × M sample matrices are normalized respectively, and in step S5, the sample matrix to be classified is also normalized, so that the normalized sample matrix and the sample matrix to be classified are adopted when the euclidean distance is calculated in steps S2 and S5,wherein element d 'of the processed matrix is normalized'hjComprises the following steps:
wherein d ishjMin (D) is the minimum value in the matrix before normalization processing; max (d) is the maximum value in the matrix before normalization.
Compared with the prior art, the invention has the beneficial effects that: according to the parallel two-finger mechanical claw provided by the invention, the plurality of elastic protrusions are respectively arranged on the two parallel clamping jaws in an array manner, and the cavities are arranged in the elastic protrusions, so that the stressed elastic protrusions can contract inwards when the two clamping jaws clamp an object, the contact area of the surfaces of the clamping jaws is increased when the mechanical claw grabs the object, and the stability and the adaptability of the mechanical claw to the grabbing of objects with different shapes are improved.
In a further scheme, the invention also has the following beneficial effects:
(1) the root of the elastic protrusion part on the clamping jaw is communicated with the air cavity, the size of air pressure can be controlled by inflating or deflating the air cavity, the elasticity of the elastic protrusion part can be further changed through the size of the air pressure, the size of the pressure when the object is grabbed can be controlled, and therefore the clamping jaw can be suitable for grabbing objects made of various materials, and grabbing of irregular and fragile objects can be achieved.
(2) The cavity in the clamping jaw can be divided into a plurality of air cavities by the air cavity partition plates, so that the stability of the air pressure in the whole system can be ensured, and the influence on the air pressure of the whole air cavity due to overlarge local pressure can be avoided.
(3) The force sensor array is arranged on the clamping jaw, and the positions of the elastic cylinders in the elastic protruding parts arranged in the array correspond to the positions of the force sensor array, so that the force generated after the elastic protruding parts are compressed is detected to detect the clamping force of the elastic protruding parts of the clamping jaw on an object, the problem of low resolution of the piezoresistor can be solved, and the force at each position can be reflected more accurately.
(4) Further on the basis of setting up the force sensor array, can train known data through the KNN classifier, further realize accomplishing the discernment to the object kind when snatching the object.
Drawings
FIG. 1 is a schematic diagram of a parallel two-finger gripper according to a preferred embodiment of the present invention;
FIG. 2 is an internal structural view of a parallel two finger gripper actuator;
FIG. 3 is a schematic view of the jaw configuration;
FIG. 4 is a schematic view of the jaw body configuration;
FIG. 5 is a schematic structural view of the rear cover;
FIG. 6 is a schematic cross-sectional view of the resilient tab;
FIG. 7 is a schematic view of the resilient protrusions in a protruding state;
FIG. 8 is a schematic view of the resilient protrusions in a compressed state;
FIG. 9 is a schematic view of a corresponding array of points of resilient protrusions in a jaw;
fig. 10 is a classification schematic diagram of the KNN classifier.
Detailed Description
In order to make the technical problems, technical solutions and advantageous effects to be solved by the embodiments of the present invention more clearly apparent, the present invention is further described in detail below with reference to the accompanying drawings and the embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
It will be understood that when an element is referred to as being "secured to" or "disposed on" another element, it can be directly on the other element or be indirectly on the other element. When an element is referred to as being "connected to" another element, it can be directly connected to the other element or be indirectly connected to the other element. In addition, the connection may be for either a fixing function or a circuit connection function.
It is to be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," and the like are used in an orientation or positional relationship indicated in the drawings for convenience in describing the embodiments of the present invention and to simplify the description, and are not intended to indicate or imply that the referenced device or element must have a particular orientation, be constructed in a particular orientation, and be in any way limiting of the present invention.
Furthermore, the terms "first", "second" and "first" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of the embodiments of the present invention, "a plurality" means two or more unless specifically limited otherwise.
The mechanical claw mainly comprises a structure, a material and a sensor, wherein the structure determines the working mode of the mechanical claw, the material determines the performance and the application of the mechanical claw, and the sensor determines the operation precision and the sensing capability of the mechanical claw. The structure, materials and sensors of the gripper are designed as follows in the preferred embodiment of the present invention.
As shown in fig. 1, the parallel two-finger gripper according to the preferred embodiment of the present invention includes a driver 10 and two parallel gripping jaws 20, the two gripping jaws 20 are connected to the driver 10, the driver 10 is used for driving at least one gripping jaw 20 to move, a plurality of elastic protrusions 211 are respectively arranged on the opposite inner sides of the two gripping jaws 20 in an array, and each elastic protrusion 211 has a cavity therein so that the elastic protrusion 211 stressed when the two gripping jaws 20 grip an object can contract inward.
In the present embodiment, the driver 10 includes a motor 11 and a gear box 12, and referring to fig. 2, the motor 11 includes a gear 111, and a jaw gear guide 121 and a connecting member 122 are disposed in the gear box 12; the gear 111 connected with the motor 11 drives the jaw gear guide rail 121 to move, the connecting piece 122 is driven to move by the movement of the jaw gear 121, and the connecting piece 122 is connected to the jaw 20, so that the closing and the opening of the two jaws are driven by the driver 10. Specifically, gear 111 that motor 11 connects sets up in the centre, the clamping jaw gear guide rail 121 that two clamping jaws 20 correspond sets up both sides around gear 111, two clamping jaw gear guide rails 121 of both sides correspond two clamping jaws 20 of control left and right sides respectively around, the rotation through gear 111 can drive the clamping jaw gear guide rail 121 of both sides around and be close to the motion relatively or keep away from the motion back to back along the direction of left and right sides, thereby realize the closure and the opening of two clamping jaws 20 of drive left and right sides, make two clamping jaws 20 can further realize the feedback of snatching and power to the object. In some embodiments, the driver 10 may be used to drive only one of the jaws 20, and the other jaw 20 may be directly and fixedly connected to the driver 10, and the driver 10 drives one of the jaws 20 to move toward or away from the other jaw 20 to close or open the two jaws.
With reference to fig. 3, 4 and 5, the clamping jaw 20 includes a clamping jaw body 21 and a back cover 22, a plurality of elastic protrusions 211 arranged in an array are arranged on the surface of the clamping jaw body 21, at least a part of the outer diameter of each elastic protrusion 211 in the direction from the top to the root is gradually increased, and with reference to fig. 6, each elastic protrusion 211 adopts a layered telescopic structure, and the inner diameter of each layer of telescopic structure is gradually increased in the direction from the top 2112 to the root 2113 of the elastic protrusion 211, so that the elastic protrusions 211 can be compressed inwards one by one when compressed. In this embodiment, the interior of the clamping jaw body 21 is further provided with air cavity partition plates 212 and air tubes 213, as shown in fig. 4, two sets of air cavity partition plates 212 divide the inner cavity space of the clamping jaw body 21 into an inner layer, a middle layer and an outer layer, three closed air cavities 23 can be formed by covering the rear cover 22, each air cavity 23 is connected with an individual air tube 213 to control the air pressure in each air cavity, wherein each elastic protrusion 211 is respectively communicated with the air cavity 23 corresponding to each position, the air tube 213 controls the stability of the air pressure in the air cavity to realize inflation and deflation, and further the elastic protrusions 211 can realize the grabbing and controlling of objects by the force of the air pressure. The rear cover 22 is provided with a force sensor array 221, the force sensor array 221 corresponds to the positions of the elastic protrusions 211 arranged in an array, and an elastic cylinder 2111 extending to the root of each elastic protrusion 211 is arranged on the inner wall (or the top of the cavity) at the top of each elastic protrusion 211, so that the end of the elastic cylinder 2111 corresponding to the elastic protrusion 211 stressed when the two clamping jaws 20 clamp an object can be in contact with the force sensor array 221, and the force of the elastic protrusion 211 contacting the plane where the force sensor array 221 is located after the object is compressed can be detected. The force sensor array 221 may be a piezoresistive force sensor array, and the length of the elastic cylinder 2111 is less than the length of the elastic protrusion 211 and greater than the sum of the length of the elastic protrusion 211 in the layer of the telescopic structure at the root and the distance from the root of the elastic protrusion 211 to the plane where the force sensor array 221 is located. The protruding state of the elastic protrusion 211 is shown in fig. 7, the compressed state of the elastic protrusion 211 is shown in fig. 8, since the thickness of each layer of the elastic protrusion 211 adopting the layered telescopic structure is different, the former layer can be compressed into the latter layer, and when the elastic protrusion 211 is contracted, the former layer can be contracted together, so that the elastic protrusion 211 has an obvious layered structure and can be restored to the protruding state by air pressure; and the elastic cylinder 2111 can contact the force sensor array in the compressed state to detect the compression force applied to the corresponding elastic protrusion 211.
The process of grabbing an object by adopting the parallel two-finger gripper of the embodiment comprises the following steps: firstly, the motor 11 drives the two clamping jaws 20 to be in an open state, all pneumatic elastic protrusions 211 are in an inflated state, when an object needs to be grabbed, the motor 11 drives the two clamping jaws 20 to move inwards relatively, the elastic protrusions 211 are compressed due to contact with the object, elastic cylinders 2111 corresponding to the compressed elastic protrusions 211 contact the force sensor array 221, the force generated by the clamping jaws 20 on the object is obtained through the force sensor array 221, and which areas of the mechanical jaws are in contact with the object can be obtained; wherein the elastic protrusion 2111 can be adapted to objects of different materials by adjusting the magnitude of the air pressure.
In this embodiment, the elastic protrusion 211 may be made of a silicone material, and the elastic cylinder 2111 may also be made of a silicone material, or a spring structure, etc.
The parallel two-finger gripper of the embodiment has the following advantages:
(1) the parallel clamping jaw structure with the large-area pneumatic elastic protruding parts arranged in the dense array is designed, the elasticity of the elastic protruding parts can be changed by controlling the air pressure of the elastic protruding parts, so that an object can be well grasped, and meanwhile, the parallel clamping jaw structure has good adaptability and grasping effect on the object with an irregular shape.
(2) The force sensor array with the piezoresistors is designed, and the positions of the elastic cylinders in the elastic protrusions arranged in the array correspond to the positions of the force sensor array, so that the force generated after each elastic protrusion is compressed is detected to determine the clamping force on an object, the problem of low resolution of the piezoresistors can be solved, and the force at each position can be reflected more accurately.
(3) The air cavity of the clamping jaw is divided into three air cavities from inside to outside in the embodiment, so that the stability of the air pressure in the whole system can be ensured, and the influence on the air pressure of the whole air cavity due to overlarge local pressure can be avoided. In other embodiments, any number of air chambers may be partitioned by the air chamber partition plate according to the number of the elastic protrusions arranged in an array in the clamping jaw, where each air chamber may be disposed as an inner layer and an outer layer in this embodiment, or may be arranged in an array, etc.
The following further describes a method for identifying the type of the object to be gripped by using the parallel two-finger gripper according to this embodiment.
When the mechanical claw grabs the object, some elastic protrusions on the surface of the mechanical claw are compressed, and the object can be identified by acquiring the position and the force for generating the compression.
And (3) identification algorithm: two 10 x 6 matrices are obtained according to the information of 60 points of the left and right clamping jaw surfaces, then the two matrices are connected into a 10 x 12 matrix, and the two matrices are placed into a KNN classifier for classification. And the object data with known objects to be grabbed is taken as a training set.
The KNN classifier is designed in this embodiment as:
(1) assuming that N types of objects are to be grabbed, each type of object is grabbed M times by using a mechanical claw,n × M data sets are obtained, each data set being a matrix of 10 × 12. In FIG. 9, the data detected at the ith point of 60 points in the left jaw is liAnd the data detected at the ith point of 60 points in the right clamping jaw is riThe resulting sample matrix is:
for ease of analysis, the above matrix is rewritten as a standard determinant, with the element in row h and column j being defined as dhj：
(2) Data preprocessing: normalizing the data in the matrix of 10-12, defining a normalized sample matrix as D ', and defining normalized elements D'hjComprises the following steps:
where min (D) represents the minimum value in the D matrix, and min (D) represents the maximum value in the D matrix.
(3) The euclidean distance Dist (x, y) is defined, M × N samples are present in the classifier, x, y represent a normalized matrix of a matrix that captures any two samples M × N times (or are converted into two N-dimensional vectors, N is the sum of the number of elastic protrusions of the two jaws, 60 × 2 ═ 120 in this embodiment), xi、yiRespectively representing the ith element of the normalized matrix of the matrix of any two corresponding samples. And (3) calculating the Euclidean distance from all the test sample points to each other sample point:
(4) sorting each euclidean distance, and classifying samples corresponding to M sample matrices with the smallest euclidean distance into one class, and classifying the samples into N classes (specifically, M × N samples, randomly selecting one sample matrix as a center, selecting M-1 sample matrices and the sample matrix as one class by taking the sample matrix as the center, and classifying the sample matrices into N classes in total) as shown in fig. 10;
(5) and (3) comparing the categories of the M points, and classifying and identifying the grabbed object by classifying the test sample points into the category with the highest ratio among the M points according to the principle that a minority obeys majority.
(6) And placing the test sample points to be classified into a KNN classifier, and selecting the class with the sample points closest to each class as the class of the sample by the classifier.
After the object is trained by the KNN classifier, the object in the trained sample can be grabbed by the mechanical claw to be classified. When gripping, the object causes the elastic protrusions to contract, and the object is classified according to the position of the elastic protrusions that contract when gripping the object and the magnitude of the force caused.
The KNN-based grabbing detection and recognition algorithm is designed in the embodiment, due to the fact that different objects are different in shape, the positions and the sizes of pressure generated on the surface of the mechanical claw are different when the mechanical claw is grabbed, the positions and the forces of elastic protruding parts, compressed when the mechanical claw is grabbed, of the different objects are collected through the piezoresistors, information is used as a classification basis, and after the algorithm is trained for multiple times, the ability of recognizing the types of the grabbed objects through touch can be achieved.
The surfaces of the parallel claws of the traditional parallel two-finger mechanical claw are mostly made of sponge or elastic silica gel, the scheme is simple and feasible, but the elastic sponge, the silica gel and other objects have small elasticity and are relatively fixed, and the problems of unstable grabbing, uneven stress and the like exist when the objects with irregular shapes are grabbed. The invention aims to design a mechanical claw with better stability and adaptability when grabbing objects of different shapes, the mechanical claw can change the elasticity of the surface of a clamping jaw by controlling air pressure, a pressure sensor below an elastic protrusion part is used for detecting the magnitude of pressure borne by the mechanical claw, and the shape of the grabbed object is identified according to the pressed position of the elastic protrusion part.
The foregoing is a more detailed description of the invention in connection with specific preferred embodiments and it is not intended that the invention be limited to these specific details. For those skilled in the art to which the invention pertains, several equivalent substitutions or obvious modifications can be made without departing from the spirit of the invention, and all the properties or uses are considered to be within the scope of the invention.
Claims (10)
1. The parallel two-finger mechanical claw is characterized by comprising a driver and two parallel clamping jaws, wherein the two clamping jaws are connected to the driver, the driver is used for driving at least one clamping jaw to move, a plurality of elastic protruding parts are arranged on the inner sides of the clamping jaws, which are arranged oppositely, in an array mode respectively, and cavities are formed in the elastic protruding parts so that the elastic protruding parts can contract inwards when the clamping jaws clamp an object under stress.
2. The parallel finger gripper according to claim 1, wherein each of said elastic protrusions has an outer diameter which is gradually increased at least in part in a direction from the tip portion to the root portion.
3. The parallel finger gripper according to claim 1, wherein each of the elastic protrusions has a layered structure, and the inner diameter of each layer of the structure increases from the top to the bottom of the elastic protrusion.
4. The parallel finger gripper according to claim 1, wherein the cavities of the elastic protrusions on each gripper are communicated with at least one air cavity, and the outlet of each air cavity is connected with an air pipe to inflate and deflate the air cavity through the air pipe.
5. The parallel finger gripper according to claim 4, wherein at least one set of air chamber partitions is provided in each of the gripping jaws to partition the cavity in the gripping jaw into a plurality of air chambers, and the cavities of the plurality of elastic protrusions are respectively communicated with the air chambers corresponding to the respective positions.
6. The parallel finger gripper according to any one of claims 1 to 5, wherein each of the gripping jaws is further provided with a force sensor array, and the position of the force sensor array corresponds to the position of the plurality of elastic protrusions arranged in an array; and an elastic cylinder extending to the root of each elastic protrusion is arranged on the inner wall of the top of each elastic protrusion, so that the end part of the elastic cylinder corresponding to the elastic protrusion stressed when the two clamping jaws clamp the object can be in contact with the force sensor array.
7. The parallel finger gripper of claim 6, wherein the length of said resilient cylinder is less than the length of said resilient protrusion.
8. The parallel finger gripper according to claim 6, wherein the force sensor array is a piezoresistive force sensor array.
9. A method for identifying the type of an object to be grabbed by using the parallel two-finger gripper as claimed in claim 6, comprising the steps of:
s1: providing N types of objects, grabbing each type of object M times by the two parallel fingers, and respectively collecting a matrix of forces borne by the force sensor arrays of the two clamping jaws to form a sample matrix when grabbing the object by the two parallel fingers each time to obtain N x M sample matrices;
s2: calculating Euclidean distance Dist (x, y) of any two sample matrixes:
wherein x and y are any two of the sample matrixes respectively, xi、yiRespectively representing the ith element of the x and y matrixes, and n is the sum of the number of the elastic protrusions of the two clamping jaws;
s3: sequencing each Euclidean distance, and dividing samples corresponding to M sample matrixes with the minimum Euclidean distance into one class and N classes in total;
s4: classifying the M samples in each class into the class with the highest proportion of the M samples to form a KNN classifier;
s5: the method comprises the steps of grabbing an object to be classified by the parallel two-finger mechanical gripper, collecting a matrix of forces borne by the force sensor arrays of the two clamping jaws to form a sample matrix to be classified, putting the sample matrix to be classified into the KNN classifier, and selecting the class closest to the Euclidean distance of the sample matrix to be classified as the class of the object to be classified by the KNN classifier.
10. The method according to claim 9, wherein before executing step S2, N × M sample matrices are further normalized, and in step S5, the sample matrix to be classified is further normalized, so that in the step S2 and step S5, the normalized sample matrix and the sample matrix to be classified are adopted when calculating euclidean distance, wherein the elements d 'of the normalized matrix'hjComprises the following steps:
wherein d ishjMin (D) is the minimum value in the matrix before normalization processing; max (d) is the maximum value in the matrix before normalization.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110290016.0A CN113043313A (en) | 2021-03-18 | 2021-03-18 | Parallel two-finger mechanical gripper and method for identifying type of gripped object by same |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110290016.0A CN113043313A (en) | 2021-03-18 | 2021-03-18 | Parallel two-finger mechanical gripper and method for identifying type of gripped object by same |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113043313A true CN113043313A (en) | 2021-06-29 |
Family
ID=76513807
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110290016.0A Pending CN113043313A (en) | 2021-03-18 | 2021-03-18 | Parallel two-finger mechanical gripper and method for identifying type of gripped object by same |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113043313A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113500614A (en) * | 2021-07-12 | 2021-10-15 | 洛阳尚奇机器人科技有限公司 | Inflatable flexible clamping jaw system |
| CN113733139A (en) * | 2021-09-30 | 2021-12-03 | 广州海洋地质调查局 | Manipulator for underwater operation |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5853211A (en) * | 1997-01-10 | 1998-12-29 | Btm Corporation | Universal gripper |
| CN105058372A (en) * | 2015-09-08 | 2015-11-18 | 清华大学 | Multi-dimensional rod group self-adaption robot grasping device |
| CN109129547A (en) * | 2018-08-29 | 2019-01-04 | 天津市远卓自动化设备制造有限公司 | A kind of automated production workshop part capture apparatus that stability is strong |
| CN209665415U (en) * | 2018-12-07 | 2019-11-22 | 广州乾通机电技术服务有限公司 | A kind of cap pneumatic picker with free buffer function |
| CN209999218U (en) * | 2019-06-04 | 2020-01-31 | 济南翼菲自动化科技有限公司 | Matrix piston gripper |
| CN209999217U (en) * | 2019-06-04 | 2020-01-31 | 济南翼菲自动化科技有限公司 | Matrix type air bag soft gripper |
| CN111283714A (en) * | 2020-03-04 | 2020-06-16 | 张梅 | Clamping jaw mechanism controlled by servo electric cylinder |
-
2021
- 2021-03-18 CN CN202110290016.0A patent/CN113043313A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US5853211A (en) * | 1997-01-10 | 1998-12-29 | Btm Corporation | Universal gripper |
| CN105058372A (en) * | 2015-09-08 | 2015-11-18 | 清华大学 | Multi-dimensional rod group self-adaption robot grasping device |
| CN109129547A (en) * | 2018-08-29 | 2019-01-04 | 天津市远卓自动化设备制造有限公司 | A kind of automated production workshop part capture apparatus that stability is strong |
| CN209665415U (en) * | 2018-12-07 | 2019-11-22 | 广州乾通机电技术服务有限公司 | A kind of cap pneumatic picker with free buffer function |
| CN209999218U (en) * | 2019-06-04 | 2020-01-31 | 济南翼菲自动化科技有限公司 | Matrix piston gripper |
| CN209999217U (en) * | 2019-06-04 | 2020-01-31 | 济南翼菲自动化科技有限公司 | Matrix type air bag soft gripper |
| CN111283714A (en) * | 2020-03-04 | 2020-06-16 | 张梅 | Clamping jaw mechanism controlled by servo electric cylinder |
Non-Patent Citations (1)
| Title |
|---|
| 严正红: "基于机器人触觉阵列信息的果蔬硬度识别研究", 《中国优秀硕士学位论文全文数据库 农业科技》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113500614A (en) * | 2021-07-12 | 2021-10-15 | 洛阳尚奇机器人科技有限公司 | Inflatable flexible clamping jaw system |
| CN113500614B (en) * | 2021-07-12 | 2023-03-14 | 洛阳尚奇机器人科技有限公司 | Inflatable flexible clamping jaw system |
| CN113733139A (en) * | 2021-09-30 | 2021-12-03 | 广州海洋地质调查局 | Manipulator for underwater operation |
| CN113733139B (en) * | 2021-09-30 | 2022-09-30 | 广州海洋地质调查局 | Manipulator for underwater operation |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Kappassov et al. | Tactile sensing in dexterous robot hands | |
| CN113043313A (en) | Parallel two-finger mechanical gripper and method for identifying type of gripped object by same | |
| Navarro et al. | Haptic object recognition for multi-fingered robot hands | |
| Shih et al. | Custom soft robotic gripper sensor skins for haptic object visualization | |
| Xie et al. | A proprioceptive soft tentacle gripper based on crosswise stretchable sensors | |
| Schill et al. | Learning continuous grasp stability for a humanoid robot hand based on tactile sensing | |
| Scimeca et al. | Model-free soft-structure reconstruction for proprioception using tactile arrays | |
| CN1757490A (en) | Hand robot | |
| JPH08323678A (en) | Soft object grip device | |
| CN113021394A (en) | Pneumatic soft multi-finger hand with cross-mode touch sensing function on palm | |
| Buescher et al. | Augmenting curved robot surfaces with soft tactile skin | |
| Takamuku et al. | Repetitive grasping with anthropomorphic skin-covered hand enables robust haptic recognition | |
| Wang et al. | Inflatable particle-jammed robotic gripper based on integration of positive pressure and partial filling | |
| Van Wyk et al. | Calibration and analysis of tactile sensors as slip detectors | |
| CN113618709A (en) | Multi-mode force-control lossless grabbing device for intelligent production line | |
| CN116348251A (en) | Interactive haptic perception method for classification and recognition of object instances | |
| Khin et al. | Development and grasp stability estimation of sensorized soft robotic hand | |
| Zheng et al. | DotView: A low-cost compact tactile sensor for pressure, shear, and torsion estimation | |
| JP4591043B2 (en) | Method of gripping an arbitrarily shaped object by a robot | |
| JP2001105378A (en) | Handling device | |
| JP2008183629A (en) | Robot and its control device and control method | |
| Cook et al. | Tactile sensor array laden 3D-printed soft robotic gripper | |
| Su et al. | Soft tactile sensing for object classification and fine grasping adjustment using a pneumatic hand with an inflatable palm | |
| Li et al. | Learning to detect slip for stable grasping | |
| Pohtongkam et al. | Object recognition using glove tactile sensor |
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: 20210629 |
|
| RJ01 | Rejection of invention patent application after publication |