CN106182054A - Finger tip vision crest line identification adaptive robot arm device - Google Patents
Finger tip vision crest line identification adaptive robot arm device Download PDFInfo
- Publication number
- CN106182054A CN106182054A CN201610640748.7A CN201610640748A CN106182054A CN 106182054 A CN106182054 A CN 106182054A CN 201610640748 A CN201610640748 A CN 201610640748A CN 106182054 A CN106182054 A CN 106182054A
- Authority
- CN
- China
- Prior art keywords
- finger
- drive
- axle
- joint shaft
- image processing
- 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.)
- Granted
Links
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
- B25J15/02—Gripping heads and other end effectors servo-actuated
- B25J15/0206—Gripping heads and other end effectors servo-actuated comprising articulated grippers
-
- 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
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B25—HAND TOOLS; PORTABLE POWER-DRIVEN TOOLS; MANIPULATORS
- B25J—MANIPULATORS; CHAMBERS PROVIDED WITH MANIPULATION DEVICES
- B25J9/00—Programme-controlled manipulators
- B25J9/16—Programme controls
- B25J9/1694—Programme controls characterised by use of sensors other than normal servo-feedback from position, speed or acceleration sensors, perception control, multi-sensor controlled systems, sensor fusion
- B25J9/1697—Vision controlled systems
Abstract
Finger tip vision crest line identification adaptive robot arm device, belong to robot vision technique field, including pedestal, three motors, drive mechanism, the first joint shaft, palm, two parallel holding fingers, two video cameras, image processing module and controller etc..The arrangement achieves the parallel holding function that robot captures orientation self-adjusting and is suitable for the situations such as object is the most placed in the middle；This device has parallel clamping grasp mode, is treated by vision-based detection and grabs object and carry out crest line identification, automatically adjusts the crawl orientation of palm and finger according to the result of crest line identification；And according to whether contact the visual feedback of object to be grabbed, automatically control the motion of different finger, thus meet the good crawl of object not placed in the middle.
Description
Technical field
The invention belongs to robot vision technique field, particularly to a kind of finger tip vision crest line identification self adaptation machine
The structure design of hand apparatus.
Background technology
Vision has been used widely as the important means of man-machine interaction, becomes current study hotspot.But by
Contain much information in vision technique, the reason such as real time handling requirement is high, Visual Feature Retrieval Process and identification difficulty, cause vision at machine
People's application aspect also has many untapped spaces.
Robot is often required to for object to adjust attitude on capturing, to be conducive to stable crawl.Exist on object
Many corner angle and crest line, be not appropriate for capturing, i.e. when finger contact corner angle or crest line, it is difficult to stably apply grasping force and cause
Capture unsuccessfully.Traditional robot does not possess the ability automatically identified corner angle or crest line, it is difficult to make suitably adjustment.
When robot captures object, object putting position is often the most unknown, there is crawl scope substantially, but does not has
There is accurate putting position, also there is no need requirement operator and the object that last operation completes is placed in rational position,
What traditional robot often required that object is placed in accurate position, if object deviation is relatively big, then causes capturing unsuccessfully.
In addition to the specialized robots such as magnetic absorption, air pressure absorption or Electrostatic Absorption, robot is mainly with two
Individual or the multifingered robot hands of multiple finger, including industry clamper, under-actuated robot hand and apery type Dextrous Hand.Industry folder
Holder is widely used in various manufacturing field, is presently the most the robot of maturation, the most all has two fingers, right
Claiming to arrange, during crawl, two fingers move in parallel, the most parallel clamping grasp mode, and this robot is referred to as parallel clamping machine
Staff.Traditional robot moved in parallel is only parts, does not have the energy of excellent perception to environment and object
Power, part has the robot of perception and realizes mainly by sense of touch, and utilizes multiple sensor to detect each pass
The position of joint, thus reach stability contorting and the control accuracy of the closed loop feedback of each joint position of finger or speed.
Visual pattern is the sensing media that a kind of quantity of information is the biggest, quickly extracts feature angle point, straight line or song from image
The technology of line has been developed that, is aided with the mode of color or priori, robot can be helped to improve and capturing work
Intelligent in work.In order to preferably capture object, traditional object recognition technique needs to carry out object at the image of complexity
Reason, including vision invariants image processing algorithm (SIFT etc.), characteristic angle point detecting method (Harris etc.), marginal point extraction side
Method (Sobel, Canney, Robert, Prewits etc.), line detection method (least square fitting, Hough, GPI etc.), erosion
Or expansion etc..Complicated algorithm is under different photoenvironments, and often reliability is the highest, and success rate ratio is relatively low, it is difficult to be suitable for
In widely collection of objects.
Summary of the invention
The invention aims to overcome the weak point of prior art, it is provided that a kind of finger tip vision crest line identification is adaptive
Answer robot hand device.This device has parallel clamping grasp mode, treats and grabs object and carry out vision-based detection, according to vision-based detection
Result adjusts crawl orientation automatically, and disclosure satisfy that the good parallel crawl of object not placed in the middle.
Technical scheme is as follows:
The present invention design a kind of finger tip vision crest line identification adaptive robot arm device, including pedestal, the first motor,
First drive mechanism, the first joint shaft, palm, the first finger and second finger；Described first motor is fixed on pedestal, described
The output shaft of the first motor and the input of the first drive mechanism are connected, the outfan of described first drive mechanism and the first joint
Axle is connected, and described first joint shaft is set in pedestal, and described palm is fixed on the first joint shaft；Described palm is respectively with
One finger, second finger are connected；Described first finger is arranged symmetrically with second finger, and its axis of symmetry is the center of the first joint shaft
Line；Described first finger, second finger are respectively provided with at least one translation joint；When capturing object, described first finger, second
Finger moves respectively along the first joint shaft vertical direction, described second finger and the first finger move toward one another；It is characterized in that:
This finger tip vision crest line identification adaptive robot arm device also includes the first video camera, the second video camera, the first image procossing
Module, the second image processing module, controller, the first driving module, the second driving module and the 3rd drive module；Described first
Finger includes the first finger base, the second motor, the first finger tips segment, the first remote joint shaft and the second drive mechanism；Institute
State second finger and include second finger pedestal, the 3rd motor, second finger end segment, the second remote joint shaft and the 3rd driver
Structure；Described second motor is fixed in the first finger base, the output shaft of described second motor and the input of the second drive mechanism
End is connected, and the outfan of described second drive mechanism and the first remote joint shaft are connected, and the described first remote joint shaft is set in first
In finger base, described first finger tips segment is socketed on the first remote joint shaft；Described first finger base is fixed in hands
Palm, described first video camera is arranged in the first finger tips segment, if the table that the first finger tips segment contacts with object
Face is Q1, and the camera lens optical axis of the first video camera is perpendicular to Q1；Described 3rd motor is fixed on second finger pedestal, and the described 3rd
The output shaft of motor and the input of the 3rd drive mechanism are connected, the outfan of described 3rd drive mechanism and the second remote joint shaft
Being connected, the described second remote joint shaft is set on second finger pedestal, and described second finger end segment is socketed in second and far closes
On nodal axisn；Described second finger pedestal is fixed on palm, and described second video camera is arranged in second finger end segment, if
The surface that second finger end segment contacts with object is Q2, and the camera lens optical axis of the second video camera is perpendicular to Q2；Described controller
Be connected with the first image processing module, the second image processing module respectively, described controller respectively with first drive module, second
Module is driven to drive module to be connected with the 3rd；The signal exit of described first video camera and the signal of the first image processing module
Input connects, and the signal exit of described second video camera and the signal input part of the second image processing module connect；Described
The side image of the first camera acquisition object to be grabbed, the opposite side image of described second camera acquisition object to be grabbed；Described
Controller performs grasping means after receiving crawl signal；Described first image processing module runs the first image processing program,
Described second image processing module runs the second image processing program；
Described first image processing program includes that order performs following steps:
(a1) shooting obtains image I11, and I11 carries out Edge extraction, obtains the result images containing marginal point
I12；
(b1) for I12, carry out crest line detection, obtain the object crest line K1 to be grabbed position on image and direction, then enter
Go next step；If not finding crest line K1 in I12, then set flag bit P1=0, perform (d1) step；
(c1) when in I12 crest line K1 through preset grabbed region H1, set flag bit P1=1；Otherwise, flag bit is set
P1=0；
(d1) calculate the average gray value T1 of I11, when T1 is less than predetermined threshold value T, then set flag bit S1=1, otherwise set
Determining flag bit S1=0, the first image processing program terminates；
Described second image processing program includes that order performs following steps:
(a2) shooting obtains image I21, and I21 carries out Edge extraction, obtains the result images containing marginal point
I22；
(b2) for I22, carry out crest line detection, obtain the object crest line K2 to be grabbed position on image and direction, then enter
Go next step；If not finding crest line K2 in I22, then set flag bit P2=0, carry out (d2) step；
(c2) when in I22 crest line K2 through preset grabbed region H2, set flag bit P2=1；Otherwise, flag bit is set
P2=0；
(d2) calculate the average gray value T2 of I21, when T2 is less than predetermined threshold value T, then set flag bit S2=1, otherwise set
Determining flag bit S2=0, the second image processing program terminates；
Described grasping means includes that order performs following steps:
(a3) perform the first image processing program and the second image processing program, obtain P1, P2, S1 and S2；
(b3) if P1=0 and P2=0, perform step (d3), otherwise perform next step；
(c3) send order-driven the first electric machine rotation low-angle, return step (a3)；
(d3) drive the first motor stalling, drive the second motor, the 3rd electric machine rotation；
(e3) if S1=1, perform next step, otherwise perform step (g3)；
(f3) order-driven the second motor stalling is sent；
(g3) if S2=1, perform next step, otherwise perform step (i3)；
(h3) order-driven the 3rd motor stalling is sent；
(i3) if S1=1, S2=1, perform step (k3), otherwise perform next step；
(j3) step (a3) is returned；
(k3) crawl terminates.
Finger tip vision crest line identification adaptive robot arm device of the present invention, it is characterised in that: described second passes
Motivation structure and the 3rd drive mechanism are the bar flat clamp mechanism of wheeled straight line；Described first finger tips segment is fixed in first and far closes
On nodal axisn；Described second finger end segment is fixed on the second remote joint shaft；The flat clamp mechanism of the wheeled straight line of described bar includes
One connecting rod, second connecting rod, third connecting rod, the first axle, the second axle, the 3rd axle, the 4th axle, the first drive, the first Flexible Transmission
Part, the second drive, the 3rd drive, the second flexible drive parts, the 4th drive；Described first axle, the second axle, the 3rd axle,
The centrage of the 4th axle and the centerline parallel of the first remote joint shaft；Described remote joint shaft is set in third connecting rod；Described
Three connecting rods are socketed on the 4th axle；The output shaft of described second motor and the first axle are connected, and described first axle sleeve is located in palm；
Described first connecting rod is fixed on the first axle；Described second axle sleeve is located on first connecting rod, one end socket of described third connecting rod
On the second axle；Described 3rd axle sleeve is located in palm, and one end of described second connecting rod is socketed on the 3rd axle, second connecting rod
The other end is socketed on the 4th axle；Described 4th axle sleeve is located at the middle part of third connecting rod；If the center of the first axle is an A, second
The center of axle is some B, and the center of the 4th axle is some C, and the center of remote joint shaft is some D, and the center of the 3rd axle is some E, line segment BC
The length three of length, the length of line segment CD and line segment CE equal, 2 times of the length of the line segment AE length equal to line segment AB,
The length of line segment CE is 2.5 times of the length of line segment AB；Described first driving wheel tube is connected on the 3rd axle, described first drive
Being fixed on palm, described first flexible drive parts connects the first drive, the second drive, described first Flexible Transmission respectively
Part is in " O " font, and described first flexible drive parts, the first drive, the second drive three constitute drive connection, and described second
Driving wheel tube is connected on the 4th axle；Described 3rd driving wheel tube is connected on the 4th axle, described second drive and the 3rd drive
Affixed, described second flexible drive parts connects the 3rd drive, the 4th drive respectively, and described second flexible drive parts is in " O "
Font, described second flexible drive parts, the 3rd drive, the 4th drive three constitute drive connection, described 4th drive
It is fixed on remote joint shaft.
Finger tip vision crest line identification adaptive robot arm device of the present invention, it is characterised in that: described first soft
Property driving member use transmission band, chain or rope, described first drive to use belt wheel, sprocket wheel or rope sheave, described second drive
Using belt wheel, sprocket wheel or rope sheave, described first flexible drive parts, the first drive, the second drive three constitute belt wheel transmission
Relation, chain gear transmission relation or rope sheave drive connection.
Finger tip vision crest line identification adaptive robot arm device of the present invention, it is characterised in that: described second soft
Property driving member use transmission band, chain or rope, described 3rd drive to use belt wheel, sprocket wheel or rope sheave, described 4th drive
Using belt wheel, sprocket wheel or rope sheave, described second flexible drive parts, the 3rd drive, the 4th drive three constitute belt wheel transmission
Relation, chain gear transmission relation or rope sheave drive connection.
The present invention compared with prior art, has the following advantages and salience effect:
Apparatus of the present invention utilize three motors, two parallel holding fingers, two video cameras, image processing module and control
Devices etc. comprehensively achieve robot finger and capture orientation self-adjusting and the parallel holding function of the situation such as to be suitable for object the most placed in the middle；
This device has parallel clamping grasp mode, is treated by vision-based detection and grabs object and carry out crest line identification, according to crest line identification
Result adjusts the crawl orientation of palm and finger automatically；And according to whether the visual feedback of contact, automatically control different finger
Motion, thus meet the good crawl of object not placed in the middle.
Accompanying drawing explanation
Fig. 1 is the solid of a kind of embodiment of the finger tip vision crest line identification adaptive robot arm device that the present invention designs
Outside drawing.
Fig. 2 is the front view of embodiment illustrated in fig. 1.
Fig. 3 is the front view (being not drawn into part) of embodiment illustrated in fig. 1.
Fig. 4 is the stereo appearance figure (being not drawn into part) of embodiment illustrated in fig. 1.
Fig. 5 is the side view (being not drawn into part) of embodiment illustrated in fig. 1.
Fig. 6 is the explosive view of embodiment illustrated in fig. 1.
Fig. 7 is single finger side outward appearance (being not drawn into part) of embodiment illustrated in fig. 1, demonstrates some A, B, C, D
Position with E.
Fig. 8 is multiple linkage sketches of AB, BC, CD, the CE shown in Fig. 6 and base linkage AE, demonstrates actively
The motion track of pivot link AB time point D, having one section of straight path in this track is exactly that the second segment straight line of the present embodiment is put down
The motion track of row reception step.
Fig. 9 is that embodiment illustrated in fig. 1 shows that robot opens up into maximum and closed configuration (double dot dash line) signal
Figure.
Figure 10 is the electrical connection diagram of embodiment illustrated in fig. 1.
Figure 11 is the flow chart of the first image processing program of embodiment illustrated in fig. 1.
Figure 12 is the flow chart of the second image processing program of embodiment illustrated in fig. 1.
Figure 13 is the grasping means flow chart of embodiment illustrated in fig. 1.
Figure 14 to Figure 17 is the course of action figure of the palm of embodiment illustrated in fig. 1.
Figure 18 to Figure 20 is the three kinds of grasp modes capturing object of embodiment illustrated in fig. 1: placed in the middle, to the left, to the right.
Figure 21 to Figure 23 be embodiment illustrated in fig. 1 identify that crest line correctly captures the course of action figure of object.
Figure 24 is that the object of embodiment illustrated in fig. 1 is can not the first video camera of crawl position or the clapped figure of the second video camera
As schematic diagram.
Figure 25 is that the object of embodiment illustrated in fig. 1 is can the first video camera of crawl position or the second video camera shot image
Schematic diagram.
In Fig. 1 to Figure 25:
1-pedestal, 11-the first motor, 12-the first drive mechanism, 13-the first joint shaft,
14-palm, 2-the first finger, 21-the first finger base, 22-the first finger tips segment,
The remote joint shaft of 23-first, 24-the second motor, 25-the second drive mechanism, 211-the first axle,
212-the second axle, 213-the 3rd axle, 214-the 4th axle, 221-first connecting rod,
222-second connecting rod, 223-third connecting rod, 231-the first drive, 232-the second drive,
233-the 3rd drive, 234-the 4th drive, 241-the first flexible drive parts, 242-second is flexible
Driving member
3-second finger, 31-second finger pedestal, 32-second finger end segment, the remote joint of 33-second
Axle,
34-the 3rd motor, 35-the 3rd drive mechanism, 41-the first video camera, 42-the second video camera,
51-the first image processing module, 52-the second image processing module, 6-controller, 71-first drives mould
Block,
72-second drives module, and 73-the 3rd drives module, 8-object, the grabbed region on H-image,
The crest line of K-object, J-viewing area.
Detailed description of the invention
Below in conjunction with the accompanying drawings and embodiment is described in further detail the content of the concrete structure of the present invention, operation principle.
A kind of embodiment of the finger tip vision crest line identification adaptive robot arm device of present invention design, such as Fig. 1 to Fig. 2
Shown in, including pedestal the 1, first motor the 11, first drive mechanism the 12, first joint shaft 13, palm the 14, first finger 2 and second
Finger 3；Described first motor 11 is fixed on pedestal 1, the output shaft of described first motor 11 and the first drive mechanism 12 defeated
Entering end to be connected, outfan and first joint shaft 13 of described first drive mechanism 12 are connected, and described first joint shaft 13 is set in
In pedestal 1, described palm 14 is fixed on the first joint shaft 13；Described palm 14 respectively with the first finger 2, second finger 3 phase
Even；Described first finger 2 is arranged symmetrically with second finger 3, and its axis of symmetry is the centrage of the first joint shaft 13；Described first-hand
Finger 2, second finger 3 are respectively provided with at least one translation joint；When capturing object 8, described first finger 2, second finger 3 are respectively
Move along the first joint shaft 13 vertical direction, described second finger 3 and the first finger 2 move toward one another；This finger tip vision crest line
Identify that adaptive robot arm device also includes first video camera the 41, second video camera the 42, first image processing module 51, second
Image processing module 52, controller 6, first drive module 71, second to drive module 72 and the 3rd to drive module 73；Described first
Finger 2 includes that the first remote joint shaft of finger base the 21, second motor the 24, first finger tips segment 22, first 23 and second passes
Motivation structure 25；Described second finger 3 includes that second finger pedestal the 31, the 3rd motor 34, second finger end segment 32, second are remote
Joint shaft 33 and the 3rd drive mechanism 35；Described second motor 24 is fixed in the first finger base 21, described second motor 24
Output shaft and the second drive mechanism 25 input be connected, the outfan of described second drive mechanism 25 and the first remote joint shaft
23 are connected, and the described first remote joint shaft 23 is set in the first finger base 21, and described first finger tips segment 22 is socketed in
On first remote joint shaft 23；Described first finger base 21 is fixed on palm 14, and described first video camera 41 is arranged on first
In finger tips segment 22, if the surface that the first finger tips segment 22 contacts with object 8 is Q1, the camera lens of the first video camera 41
Optical axis is perpendicular to Q1；Described 3rd motor 34 is fixed on second finger pedestal 31, the output shaft of described 3rd motor 34 and
The input of three drive mechanisms 35 is connected, and the outfan of described 3rd drive mechanism 35 and the second remote joint shaft 33 are connected, described
Second remote joint shaft 33 is set on second finger pedestal 31, and described second finger end segment 32 is socketed in the second remote joint shaft
On 33；Described second finger pedestal 31 is fixed on palm 14, and described second video camera 42 is arranged on second finger end segment
On 32, if the surface that second finger end segment 32 contacts with object 8 is Q2, the camera lens optical axis of the second video camera 42 is perpendicular to
Q2；Described controller 6 is connected with first image processing module the 51, second image processing module 52 respectively, as shown in Figure 10, described
Controller 6 drives module 71, second to drive module 72 and the 3rd to drive module 73 to be connected respectively with first；Described first video camera
The signal exit of 41 and the signal input part of the first image processing module 51 connect, and the signal of described second video camera 42 is drawn
End is connected with the signal input part of the second image processing module 52；Described first video camera 41 gathers the side figure of object 8 to be grabbed
Picture, described second video camera 42 gathers the opposite side image of object 8 to be grabbed；Described controller 6 performs after receiving crawl signal
Grasping means；Described first image processing module 51 runs the first image processing program, and described second image processing module 52 is transported
Row the second image processing program；
Described first image processing program includes that order performs following steps:
(a1) shooting obtains image I11, and I11 carries out Edge extraction, obtains the result images containing marginal point
I12；
(b1) for I12, carry out crest line detection, obtain the object 8 crest line K1 to be grabbed position on image and direction, then enter
Go next step；If not finding crest line K1 in I12, then set flag bit P1=0, perform (d1) step；
(c1) when in I12 crest line K1 through preset grabbed region H1, set flag bit P1=1；Otherwise, flag bit is set
P1=0；
(d1) calculate the average gray value T1 of I11, when T1 is less than predetermined threshold value T, then set flag bit S1=1, otherwise set
Determining flag bit S1=0, the first image processing program terminates；
Described second image processing program includes that order performs following steps:
(a2) shooting obtains image I21, and I21 carries out Edge extraction, obtains the result images containing marginal point
I22；
(b2) for I22, carry out crest line detection, obtain the object 8 crest line K2 to be grabbed position on image and direction, then enter
Go next step；If not finding crest line K2 in I22, then set flag bit P2=0, perform (d2) step；
(c2) when in I22 crest line K2 through preset grabbed region H2, set flag bit P2=1；Otherwise, flag bit is set
P2=0；
(d2) calculate the average gray value T2 of I21, when T2 is less than predetermined threshold value T, then set flag bit S2=1, otherwise set
Determining flag bit S2=0, the second image processing program terminates；
Described grasping means includes that order performs following steps:
(a3) perform the first image processing program and the second image processing program, obtain P1, P2, S1 and S2；
(b3) if P1=0 and P2=0, perform step (d3), otherwise perform next step；
(c3) send order-driven the first motor 11 and rotate low-angle, return step (a3)；
(d3) drive the first motor 11 stall, drive the second motor the 24, the 3rd motor 34 to rotate；
(e3) if S1=1, perform next step, otherwise perform step (g3)；
(f3) order-driven the second motor 24 stall is sent；
(g3) if S2=1, perform next step, otherwise perform step (i3)；
(h3) order-driven the 3rd motor 34 stall is sent；
(i3) if S1=1, S2=1, perform step (k3), otherwise perform next step；
(j3) step (a3) is returned；
(k3) crawl terminates.
In the present embodiment, described second drive mechanism 25 and the 3rd drive mechanism 35 are the bar flat clamp mechanism of wheeled straight line；
Described first finger tips segment 22 is fixed on the first remote joint shaft 23；Described second finger end segment 32 is fixed in second
On remote joint shaft 33；The flat clamp mechanism of the wheeled straight line of described bar include first connecting rod 221, second connecting rod 222, third connecting rod 223,
One axle the 211, second axle the 212, the 3rd axle the 213, the 4th axle the 214, first drive the 231, first flexible drive parts 241, second passes
Driving wheel the 232, the 3rd drive the 233, second flexible drive parts the 242, the 4th drive 234；Described first axle the 211, second axle
212, the centrage of the 3rd axle the 213, the 4th axle 214 and the centerline parallel of the first remote joint shaft 23；Described remote joint shaft is sheathed
In third connecting rod 223；Described third connecting rod 223 is socketed on the 4th axle 214；The output shaft and first of described second motor 24
Axle 211 is connected, and described first axle 211 is set in palm 14；Described first connecting rod 221 is fixed on the first axle 211；Described
Two axles 212 are set on first connecting rod 221, and one end of described third connecting rod 223 is socketed on the second axle 212；Described 3rd axle
213 are set in palm 14, and one end of described second connecting rod 222 is socketed on the 3rd axle 213, the other end of second connecting rod 222
It is socketed on the 4th axle 214；Described 4th axle 214 is set in the middle part of third connecting rod 223；If the center of the first axle 211 is a little
A, the center of the second axle 212 is some B, and the center of the 4th axle 214 is some C, and the center of remote joint shaft is some D, in the 3rd axle 213
The heart is an E, and as shown in Figure 7 and Figure 8, the length three of the length of line segment BC, the length of line segment CD and line segment CE is equal, line segment AE
The length 2 times of length equal to line segment AB, the length of line segment CE is 2.5 times of the length of line segment AB；Described first drive
231 are socketed on the 3rd axle 213, and described first drive 231 is fixed on palm 14, described first flexible drive parts 241 points
Not connecting first drive the 231, second drive 232, described first flexible drive parts 241 is in " O " font, and described first is flexible
Driving member the 241, first drive the 231, second drive 232 three constitutes drive connection, and described second drive 232 is socketed in
On 4th axle 214；Described 3rd drive 233 is socketed on the 4th axle 214, described second drive 232 and the 3rd drive
233 is affixed, and described second flexible drive parts 242 connects the 3rd drive the 233, the 4th drive 234 respectively, and described second is flexible
Driving member 242 is in " O " font, and described second flexible drive parts the 242, the 3rd drive the 233, the 4th drive 234 three is constituted
Drive connection, described 4th drive 234 is fixed on remote joint shaft；As shown in Figures 3 to 8.Fig. 9 is the parallel folder of the present embodiment
The original state held and final state (double dot dash line).
Finger tip vision crest line identification adaptive robot arm device of the present invention, it is characterised in that: described first soft
Property driving member 241 uses transmission band, chain or rope, described first drive 231 to use belt wheel, sprocket wheel or rope sheave, and described second
Drive 232 uses belt wheel, sprocket wheel or rope sheave, described first flexible drive parts the 241, first drive the 231, second drive
232 threes constitute belt wheel transmission relation, chain gear transmission relation or rope sheave drive connection.In the present embodiment, the described first flexible biography
Moving part 241 uses transmission band, described first drive 231 to use belt wheel, described second drive 232 to use belt wheel, and described
One flexible drive parts the 241, first drive the 231, second drive 232 three constitutes belt wheel transmission relation.
Finger tip vision crest line identification adaptive robot arm device of the present invention, it is characterised in that: described second soft
Property driving member 242 uses transmission band, chain or rope, described 3rd drive 233 to use belt wheel, sprocket wheel or rope sheave, and the described 4th
Drive 234 uses belt wheel, sprocket wheel or rope sheave, described second flexible drive parts the 242, the 3rd drive the 233, the 4th drive
234 threes constitute belt wheel transmission relation, chain gear transmission relation or rope sheave drive connection.In the present embodiment, the described second flexible biography
Moving part 242 uses transmission band, described 3rd drive 233 to use belt wheel, described 4th drive 234 to use belt wheel, and described
Two flexible drive parts the 242, the 3rd drive the 233, the 4th drive 234 threes constitute belt wheel transmission relation.
Course of action when the present embodiment rotates the first joint shaft is as shown in Figure 14 to Figure 17.
The situation captured when object is placed on to the left, to the right or middle is as shown in Figure 18 to Figure 20.
During grasping body, crest line observed can capture area H time situation as shown in figure 21.
During grasping body, after have rotated the first joint shaft according to crest line image recognition situation, reach the orientation that can capture
Time as shown in figure 22, after crawl as shown in figure 23.
The operation principle of the present embodiment, is described below in conjunction with accompanying drawing:
Being placed on crawl workbench by object 8 to be grabbed, controller 6 performs the program of grasping means, such as Figure 11 to Figure 13
Shown in, figure clapped by the first video camera 41, feeds back to the first image processing module 51, performs the first image processing program, obtains first
Whether the gripping surface of finger 2 end segment correspondence object 8 exists crest line, if there is crest line and crest line is in and can grab region H1, then
Drive the first motor 11 to rotate a low-angle, then shoot image；Second video camera 42 bat figure and the first video camera 41 are clapped figure and are processed
Program is similar.Until not there is not rib in respective region of grabbing in the image of the first video camera 41 and the second video camera 42 feedback
Line, then enter next link；
The second motor 24 and the 3rd motor 34 is driven to rotate respectively；
Figure clapped by first video camera 41, feeds back to the first image processing module 51, performs the first image processing program, and detection should
The average gray value T1 of image, if T1 is less than predetermined threshold value T, then stops the second motor 24；
Figure clapped by second video camera 42, feeds back to the second image processing module 52, performs the second image processing program, and detection should
The average gray value T2 of image, if T2 is less than predetermined threshold value T, then stops the 3rd motor 34；
If T1 is less than predetermined threshold value T less than predetermined threshold value T and T2, crawl terminates.
When controller 6 receives release signal, perform release procedure, release procedure for drive respectively the second motor 24, the
Three motors 34 are inverted to initial position.
Observed region is J, as it is shown in figure 1, the image photographed is such as Figure 24, shown in 25, wherein, image is observed rib
Line can capture area time as shown in figure 24, when crest line not can capture area time as shown in figure 25.
Described edge extracting method uses Sobel method, for the known technology of robot visual field, repeats no more.
Described crest line detection method uses line detection method, in the marginal point result images that i.e. detection edge extracting obtains
Through a most straight line, known line detection method can be used, such as Hough method or Gray Projection integration
(GPI) method, list of references is:
[1] Zheng Zhewei, Zhang Wenzeng etc. the circular subwindow multistep GPI method of image weld seam detection. welding journal .2007,
28(8):77-80.
[2] Zhang Wenzeng, Chen Qiang etc. the gray projecting integral method of straight-line detection. Tsing-Hua University journal .2005,45 (11):
1446-1449.
Apparatus of the present invention are only for the grasping body of uniform background environment, and the object captured exists crest line or do not exists
Crest line is the most applicable.Described on image can capture area H be that the parameter such as the lens focus according to video camera, crawl scope sets in advance
Fixed fixed area.Described image averaging gray value T is also fixed value set in advance.
Apparatus of the present invention utilize three motors, two parallel holding fingers, two video cameras, image processing module and control
Devices etc. comprehensively achieve robot finger and capture orientation self-adjusting and the parallel holding function of the situation such as to be suitable for object the most placed in the middle；
This device has parallel clamping grasp mode, is treated by vision-based detection and grabs object and carry out crest line identification, according to crest line identification
Result adjusts the crawl orientation of palm and finger automatically；And according to whether the visual feedback of contact, automatically control different finger
Motion, thus meet the good crawl of object not placed in the middle.
Claims (4)
1. a finger tip vision crest line identification adaptive robot arm device, including pedestal, the first motor, the first drive mechanism,
First joint shaft, palm, the first finger and second finger；Described first motor is fixed on pedestal, described first motor defeated
The input of shaft and the first drive mechanism is connected, and the outfan of described first drive mechanism and the first joint shaft are connected, described
First joint shaft is set in pedestal, and described palm is fixed on the first joint shaft；Described palm respectively with the first finger, second
Finger is connected；Described first finger is arranged symmetrically with second finger, and its axis of symmetry is the centrage of the first joint shaft；Described first
Finger, second finger are respectively provided with at least one translation joint；When capturing object, described first finger, second finger edge respectively
The first joint shaft vertical direction to move, described second finger and the first finger move toward one another；It is characterized in that: this finger tip vision
Crest line identification adaptive robot arm device also includes the first video camera, the second video camera, the first image processing module, the second figure
As processing module, controller, the first driving module, the second driving module and the 3rd drive module；Described first finger includes
One finger base, the second motor, the first finger tips segment, the first remote joint shaft and the second drive mechanism；Described second finger
Including second finger pedestal, the 3rd motor, second finger end segment, the second remote joint shaft and the 3rd drive mechanism；Described
Two motors are fixed in the first finger base, and the output shaft of described second motor and the input of the second drive mechanism are connected, institute
Outfan and the first remote joint shaft of stating the second drive mechanism are connected, and the described first remote joint shaft is set in the first finger base
On, described first finger tips segment is socketed on the first remote joint shaft；Described first finger base is fixed on palm, described
First video camera is arranged in the first finger tips segment, if the surface that the first finger tips segment contacts with object is Q1, the
The camera lens optical axis of one video camera is perpendicular to Q1；Described 3rd motor is fixed on second finger pedestal, described 3rd motor defeated
The input of shaft and the 3rd drive mechanism is connected, and the outfan of described 3rd drive mechanism and the second remote joint shaft are connected, institute
Stating the second remote joint shaft to be set on second finger pedestal, described second finger end segment is socketed on the second remote joint shaft；
Described second finger pedestal is fixed on palm, and described second video camera is arranged in second finger end segment, if second-hand
Referring to that the surface that end segment contacts with object is Q2, the camera lens optical axis of the second video camera is perpendicular to Q2；Described controller respectively with
First image processing module, the second image processing module are connected；Described controller drives module, the second driving mould respectively with first
Block and the 3rd drives module to be connected；The signal exit of described first video camera and the signal input part of the first image processing module
Connecting, the signal exit of described second video camera and the signal input part of the second image processing module connect；Described first takes the photograph
Camera gathers the side image of object to be grabbed, the opposite side image of described second camera acquisition object to be grabbed；Described controller
Grasping means is performed after receiving crawl signal；Described first image processing module runs the first image processing program, and described
Two image processing modules run the second image processing program；
Described first image processing program includes that order performs following steps:
(a1) shooting obtains image I11, and I11 carries out Edge extraction, obtains the result images I12 containing marginal point；
(b1) for I12, carry out crest line detection, obtain the object crest line K1 to be grabbed position on image and direction, then under carrying out
One step；If not finding crest line K1 in I12, then set flag bit P1=0, perform (d1) step；
(c1) when in I12 crest line K1 through preset grabbed region H1, set flag bit P1=1；Otherwise, flag bit P1=is set
0；
(d1) calculate the average gray value T1 of I11, when T1 is less than predetermined threshold value T, then set flag bit S1=1, otherwise set mark
Will position S1=0, the first image processing program terminates；
Described second image processing program includes that order performs following steps:
(a2) shooting obtains image I21, and I21 carries out Edge extraction, obtains the result images I22 containing marginal point；
(b2) for I22, carry out crest line detection, obtain the object crest line K2 to be grabbed position on image and direction, then under carrying out
One step；If not finding crest line K2 in I22, then set flag bit P2=0, perform (d2) step；
(c2) when in I22 crest line K2 through preset grabbed region H2, set flag bit P2=1；Otherwise, flag bit P2=is set
0；
(d2) calculate the average gray value T2 of I21, when T2 is less than predetermined threshold value T, then set flag bit S2=1, otherwise set mark
Will position S2=0, the second image processing program terminates；
Described grasping means includes that order performs following steps:
(a3) perform the first image processing program and the second image processing program, obtain P1, P2, S1 and S2；
(b3) if P1=0 and P2=0, perform step (d3), otherwise perform next step；
(c3) send order-driven the first electric machine rotation low-angle, return step (a3)；
(d3) drive the first motor stalling, drive the second motor, the 3rd electric machine rotation；
(e3) if S1=1, perform next step, otherwise perform step (g3)；
(f3) order-driven the second motor stalling is sent；
(g3) if S2=1, perform next step, otherwise perform step (i3)；
(h3) order-driven the 3rd motor stalling is sent；
(i3) if S1=1, S2=1, perform step (k3), otherwise perform next step；
(j3) step (a3) is returned；
(k3) crawl terminates.
2. finger tip vision crest line identification adaptive robot arm device as claimed in claim 1, it is characterised in that: described second
Drive mechanism and the 3rd drive mechanism are the bar flat clamp mechanism of wheeled straight line；It is remote that described first finger tips segment is fixed in first
On joint shaft；Described second finger end segment is fixed on the second remote joint shaft；The flat clamp mechanism of the wheeled straight line of described bar includes
First connecting rod, second connecting rod, third connecting rod, the first axle, the second axle, the 3rd axle, the 4th axle, the first drive, the first flexible biography
Moving part, the second drive, the 3rd drive, the second flexible drive parts, the 4th drive；Described first axle, the second axle, the 3rd
The centerline parallel of the remote joint shaft of axle, the centrage of the 4th axle and first；Described remote joint shaft is set in third connecting rod；Described
Third connecting rod is socketed on the 4th axle；The output shaft of described second motor and the first axle are connected, and described first axle sleeve is located at palm
In；Described first connecting rod is fixed on the first axle；Described second axle sleeve is located on first connecting rod, one end set of described third connecting rod
It is connected on the second axle；Described 3rd axle sleeve is located in palm, and one end of described second connecting rod is socketed on the 3rd axle, second connecting rod
The other end be socketed on the 4th axle；Described 4th axle sleeve is located at the middle part of third connecting rod；If the center of the first axle is an A, the
The center of two axles is some B, and the center of the 4th axle is some C, and the center of remote joint shaft is some D, and the center of the 3rd axle is some E, line segment
The length three of the length of BC, the length of line segment CD and line segment CE is equal, the 2 of the length of the line segment AE length equal to line segment AB
Times, the length of line segment CE is 2.5 times of the length of line segment AB；Described first driving wheel tube is connected on the 3rd axle, and described first passes
Driving wheel is fixed on palm, and described first flexible drive parts connects the first drive, the second drive respectively, and described first is flexible
Driving member is in " O " font, and described first flexible drive parts, the first drive, the second drive three constitute drive connection, described
Second driving wheel tube is connected on the 4th axle；Described 3rd driving wheel tube is connected on the 4th axle, and described second drive and the 3rd passes
Driving wheel is affixed, and described second flexible drive parts connects the 3rd drive, the 4th drive respectively, described second flexible drive parts in
" O " font, described second flexible drive parts, the 3rd drive, the 4th drive three constitute drive connection, described 4th transmission
Wheel is fixed on remote joint shaft.
3. finger tip vision crest line identification adaptive robot arm device as claimed in claim 2, it is characterised in that: described first
Flexible drive parts uses transmission band, chain or rope, described first drive to use belt wheel, sprocket wheel or rope sheave, described second transmission
Wheel uses belt wheel, sprocket wheel or rope sheave, described first flexible drive parts, the first drive, the second drive three to constitute belt wheel and pass
Dynamic relation, chain gear transmission relation or rope sheave drive connection.
4. finger tip vision crest line identification adaptive robot arm device as claimed in claim 2, it is characterised in that: described second
Flexible drive parts uses transmission band, chain or rope, described 3rd drive to use belt wheel, sprocket wheel or rope sheave, described 4th transmission
Wheel uses belt wheel, sprocket wheel or rope sheave, described second flexible drive parts, the 3rd drive, the 4th drive three to constitute belt wheel and pass
Dynamic relation, chain gear transmission relation or rope sheave drive connection.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610640748.7A CN106182054B (en) | 2016-08-05 | 2016-08-05 | Finger tip vision crest line identifies adaptive robot arm device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610640748.7A CN106182054B (en) | 2016-08-05 | 2016-08-05 | Finger tip vision crest line identifies adaptive robot arm device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106182054A true CN106182054A (en) | 2016-12-07 |
| CN106182054B CN106182054B (en) | 2018-11-30 |
Family
ID=57514424
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610640748.7A Active CN106182054B (en) | 2016-08-05 | 2016-08-05 | Finger tip vision crest line identifies adaptive robot arm device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106182054B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106829515A (en) * | 2017-03-31 | 2017-06-13 | 江南大学 | Using the not only extensible but also retractable palletizing mechanical arm of linkage |
| CN108145731A (en) * | 2017-11-28 | 2018-06-12 | 浙江卓怡纺织有限公司 | A kind of feeding device for tubing |
| CN112757334A (en) * | 2020-12-28 | 2021-05-07 | 清华大学 | Robot gripper and control method thereof |
| CN113291790A (en) * | 2021-05-25 | 2021-08-24 | 武汉理工大学 | Self-adaptive grabbing mechanism |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102259340A (en) * | 2011-07-11 | 2011-11-30 | 浙江理工大学 | Underactuated end effector |
| WO2013075245A1 (en) * | 2011-11-25 | 2013-05-30 | Robotiq Inc. | A gripper having a two degree of freedom underactuated mechanical finger for encompassing and pinch grasping |
| CN103395067A (en) * | 2013-07-31 | 2013-11-20 | 北京航空航天大学 | Self-adaption claw mechanism of spatial on-orbit service robot |
| CN104972475A (en) * | 2015-07-29 | 2015-10-14 | 苏州拓德机器人科技有限公司 | Under-actuated mechanical hand with self-adaptive shape |
-
2016
- 2016-08-05 CN CN201610640748.7A patent/CN106182054B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102259340A (en) * | 2011-07-11 | 2011-11-30 | 浙江理工大学 | Underactuated end effector |
| WO2013075245A1 (en) * | 2011-11-25 | 2013-05-30 | Robotiq Inc. | A gripper having a two degree of freedom underactuated mechanical finger for encompassing and pinch grasping |
| CN103395067A (en) * | 2013-07-31 | 2013-11-20 | 北京航空航天大学 | Self-adaption claw mechanism of spatial on-orbit service robot |
| CN104972475A (en) * | 2015-07-29 | 2015-10-14 | 苏州拓德机器人科技有限公司 | Under-actuated mechanical hand with self-adaptive shape |
Non-Patent Citations (4)
| Title |
|---|
| 张文增等: "拟人机器人手多指欠驱动机构研究", 《机械设计与研究》 * |
| 张文增等: "末端强力抓取的欠驱动拟人机器人手", 《清华大学学报》 * |
| 张文增等: "直线检测的灰度投影积分方法", 《清华大学学报》 * |
| 郑哲伟: "图像焊缝检测的圆形子窗口多步GPI方法", 《焊接学报》 * |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106829515A (en) * | 2017-03-31 | 2017-06-13 | 江南大学 | Using the not only extensible but also retractable palletizing mechanical arm of linkage |
| CN108145731A (en) * | 2017-11-28 | 2018-06-12 | 浙江卓怡纺织有限公司 | A kind of feeding device for tubing |
| CN112757334A (en) * | 2020-12-28 | 2021-05-07 | 清华大学 | Robot gripper and control method thereof |
| CN113291790A (en) * | 2021-05-25 | 2021-08-24 | 武汉理工大学 | Self-adaptive grabbing mechanism |
| CN113291790B (en) * | 2021-05-25 | 2022-08-30 | 武汉理工大学 | Self-adaptive grabbing mechanism |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106182054B (en) | 2018-11-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106182054A (en) | Finger tip vision crest line identification adaptive robot arm device | |
| US10913151B1 (en) | Object hand-over between robot and actor | |
| CN109465840B (en) | Sucking disc-paw composite grabbing device | |
| CN104786865B (en) | A kind of method of docking of being charged for electric automobile is provided based on monocular vision | |
| JP6823008B2 (en) | Robot system for taking out workpieces stacked separately and control method for robot system | |
| US20170181341A1 (en) | Working apparatus and working method | |
| CN108789414A (en) | Intelligent machine arm system based on three-dimensional machine vision and its control method | |
| CN107009358A (en) | A kind of unordered grabbing device of robot based on one camera and method | |
| CN206764767U (en) | The manipulator that a kind of view-based access control model captures automatically | |
| JP7044047B2 (en) | robot | |
| CN106956263A (en) | Automobile component vision guide handgrip | |
| CN110842890B (en) | Robot and control method thereof | |
| CN107649406A (en) | A kind of efficient more material picking up system of binocular vision and method | |
| CN111483803A (en) | Control method, capture system and storage medium | |
| CN114770461B (en) | Mobile robot based on monocular vision and automatic grabbing method thereof | |
| Chapman et al. | A locally-adaptive, parallel-jaw gripper with clamping and rolling capable, soft fingertips for fine manipulation of flexible flat cables | |
| Ali et al. | An intelligent adjustable spanner for automated engagement with multi-diameter bolts/nuts during tightening/loosening process using vision system and fuzzy logic | |
| CN207058571U (en) | Automobile component vision guide handgrip | |
| JP2006130580A (en) | Method for gripping arbitrarily-shaped object by robot | |
| Lin et al. | Vision based object grasping of industrial manipulator | |
| CN208020198U (en) | A kind of intelligent grabbing robot based on RGBD | |
| CN207148418U (en) | With the endoscope for salvaging function | |
| CN206392447U (en) | The detection of PE bottles and sorting system based on machine vision | |
| Bdiwi et al. | Handing-over model-free objects to human hand with the help of vision/force robot control | |
| CN206748400U (en) | A kind of industrial carrying machine people based on autonomous learning |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |