CN107972029A - Mechanical arm instructs control system - Google Patents
Mechanical arm instructs control system Download PDFInfo
- Publication number
- CN107972029A CN107972029A CN201711082492.3A CN201711082492A CN107972029A CN 107972029 A CN107972029 A CN 107972029A CN 201711082492 A CN201711082492 A CN 201711082492A CN 107972029 A CN107972029 A CN 107972029A
- Authority
- CN
- China
- Prior art keywords
- mechanical arm
- path
- motor
- dynamic
- module
- 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.)
- Withdrawn
Links
Classifications
-
- 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
- B25J9/1602—Programme controls characterised by the control system, structure, architecture
- B25J9/161—Hardware, e.g. neural networks, fuzzy logic, interfaces, processor
-
- 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
- B25J9/1656—Programme controls characterised by programming, planning systems for manipulators
- B25J9/1664—Programme controls characterised by programming, planning systems for manipulators characterised by motion, path, trajectory planning
Landscapes
- Engineering & Computer Science (AREA)
- Mechanical Engineering (AREA)
- Automation & Control Theory (AREA)
- Robotics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Fuzzy Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- Feedback Control In General (AREA)
- Numerical Control (AREA)
- Manipulator (AREA)
Abstract
The invention discloses mechanical arm to instruct control system, include operator interface device, control device and mechanical arm, operator interface device is connected with the control device, operator interface device is controlling the control device, control device is connected with mechanical arm, mechanical arm includes an at least motor driving part and an end effector, motor driving part is to driving manipulator arm, control system includes dynamic learning module, instruct memory module, path optimization's module and path execution module, the learning dynamics model produced by dynamic learning module, memory module is instructed to produce the path after mechanical arm study, with the adjustment of path optimization's module, so that path execution module accurately performs the required moving line of user.There is the study of mechanical arm dynamic model and compensation, user can directly be instructed, and be not required to additional sensor, without the use of teaching box, reduce production cost.
Description
Technical field
Instruct control system the present invention relates to a kind of mechanical arm, more particularly to it is a kind of can self-teaching mechanical arm it is dynamic
States model and the mechanical arm teaching control system that the data of dynamic model can be done to feedback compensation.
Background technology
At present, mechanical arm is widely used in industrial quarters, such as：Pick and place, cut, processing, welding, assembling etc. items
Work, can improve the degree of automation and work efficiency of production line.Mechanical arm instructs system, generate the order of mechanical arm with
Perform the work being endowed.The most common method taught of present industrial machinery arm, uses teaching box (instruction box)
Produce and be intended to the instruction of in-position point and generate path, action for compound action or to process stereo structure irregular
Workpiece, and consuming time cost more not humane using the method for teaching box, in order to solve the problem above-mentioned, without using teaching box
Mechanical arm method taught be suggested, three teachings device described briefly below, and its method that three each uses：
The first teaching system, teaching mode are manually to drive robotic arm to complete teaching path, and the method is being imparted knowledge to students
When, the motor of robotic arm must be in servo closed mode, therefore when user directly drives mechanical arm, it is necessary to bears machine
The various mechanical forces of device arm, therefore the difficulty in teaching is caused, and the transmission mechanism of high transmission ratio is also easily in the process of teaching
Middle damage.So using the made mechanical arm of this teaching mode, its weight is light and handy with build, and heavy-duty machinery arm leads to
Often without using the method.
Second of teaching system, includes a teaching mechanical arm and an execution mechanical arm, inside teaching manipulator
Arm only assembles encoder, has no assembling motor.When user drives this teaching mechanical arm, the encoder on mechanical arm is instructed
Position signal can be recorded and be converted into performing the track archives that can be read of mechanical arm, hold execution mechanical arm
This track archives of row act to complete identical teaching.The method need to additionally use teaching mechanical arm, therefore production cost meeting
Improve, and instruct mechanical arm must with perform mechanical arm mechanism it is consistent, teaching mechanical arm be not used to various configuration or
The execution mechanical arm of mechanism parameter, so the teaching mode application range of this type is relatively limited to.
The third teaching system with manually drive robotic arm complete teaching path, this mechanical arm teaching system need by
The sensors such as force snesor, accelerometer detect external force, are performed using outer force information after measurement to assign mechanical arm end
Device amount of movement.The teaching point position of the teaching mode is fixed, and by taking six axle powers rule as an example, initiator must just grasp six axle powers rule
The specific location at place is instructed, remaining position can not then impart knowledge to students.In addition, this method taught needs more assembling sensings
Device, therefore the production cost increases.
The content of the invention
The purpose of the present invention is overcome the shortcomings of the prior art, there is provided a kind of mechanical arm instructs control system, makes
User can directly be instructed, and be not required to additional sensor, without the use of teaching box, reduce production cost.
The purpose of the present invention is achieved through the following technical solutions：
Mechanical arm instructs control system, and feature is：Include operator interface device, control device and mechanical arm, institute
State operator interface device to be connected with the control device, the operator interface device is controlling the control device, the control
Device processed is connected with mechanical arm, and the mechanical arm includes an at least motor driving part and an end effector, the motor
Drive division is to driving manipulator arm；
The control device includes dynamic learning module, teaching memory module and path execution module,
The dynamic learning module, including：Digital independent portion, to the data setting file of read control device；
First order generating unit, study locus model is established according to data setting file；
First order output section, according to study locus model, calculates the position command of each motor and exports to horse
Up to drive division, wherein motor driving part produces a motor torsional moment feedback according to the position command of each motor；
Dynamic model calculating part, the learning dynamics mould of mechanical arm is obtained according to study locus model and motor torsional moment feedback
Type；
Dynamic model memory portion, to store the learning dynamics model of mechanical arm；
The teaching memory module, comprising dynamic compensating unit and mechanism converting unit,
Dynamic compensating unit, including：Dynamic model reading part, to read the machinery being stored in dynamic model memory portion
The learning dynamics model of arm；
Torque command calculating part, the learning dynamics model and motor of the mechanical arm according to stored by dynamic model memory portion
The motor position signal that drive division is measured forms a torque command signal；
Torque command output section, receives the torque command signal that is formed of torque command calculating part and to produce torque command defeated
Go out to motor driving part；The torque command signal that wherein motor driving part processing is obtained by torque command output section drives machinery
Arm, and motor position signal is generated, the processing of torque command calculating part is back to, is compensated with forming the dynamic of a loop circuit；
Mechanism converting unit, including：Mechanism changes calculating part, according to motor position feedback signal and the mechanism of mechanical arm
Parameter calculates the first position point of the end effector of mechanical arm；
The first position point of the end effector of mechanical arm, is stored in the operation road for mechanical arm by path memory portion
Footpath；
The path execution module, including：Path reading part, to read the mechanical arm for being stored in path memory portion
Courses of action；
Second order generating unit, according to the position of a perform track curve of execution mechanical arm and end effector in terms of
Calculate the position command of each motor；
Second order output section, the position command of each motor of the second order generating unit generation is exported to motor and is driven
Portion, the data that mechanical arm exports the second order output section perform courses of action with driving manipulator arm.
Further, above-mentioned mechanical arm teaching control system, wherein, the control device also includes path optimization's mould
Block, path optimization's module, including：Path reading part, to read the operation for the mechanical arm for being stored in path memory portion
Path；
Portion of path optimization, at least one miscellaneous point is to optimize courses of action in the courses of action to filtration machinery arm；
Path optimizing memory portion, the courses of action optimized to be converted into the processing shelves of mechanical arm.
Further, above-mentioned mechanical arm teaching control system, wherein, the portion of path optimization includes an Automatic Optimal
Portion and an artificial optimization portion.
Further, above-mentioned mechanical arm teaching control system, wherein, the portion of path optimization according to courses of action with
One surrounding environment, a position relative relation of a workpieces processing carry out the action of the courses of action of simulation manipulator arm to judge to be
The operation of no adjustment mechanical arm.
Further, above-mentioned mechanical arm teaching control system, wherein, the path artificial optimization portion adjusts manipulator
The operation of arm includes one smooth trajectory degree of setting and/or a profile errors.
Further, above-mentioned mechanical arm teaching control system, wherein, the position command of each motor passes through dynamics
Practise module and be sent to mechanical arm via motor driving part.
Further, above-mentioned mechanical arm teaching control system, wherein, motor torsional moment feedback is passed by motor driving part
Send to the dynamic model calculating part of dynamic learning module.
Further, above-mentioned mechanical arm teaching control system, wherein, the control device also includes one to optimize
Path optimization's module of mechanical arm courses of action.
Further, above-mentioned mechanical arm teaching control system, wherein, path optimization's module includes：Read in path
Portion is taken, to read the courses of action of mechanical arm executed；
The excellent dynamicization portion in path, to filter at least one miscellaneous point of the courses of action of tool arm executed to optimize executed
Courses of action；
Path optimizing memory portion, the courses of action of the executed optimized to be converted into a processing of mechanical arm
Shelves.
Further, above-mentioned mechanical arm teaching control system, wherein, the torque command of each motor is mended by dynamic
Repay unit and be sent to mechanical arm via motor driving part, motor position signal is sent to dynamic by motor driving part and compensates list
The torque command calculating part of member is compensated with forming loop circuit dynamic.
The present invention has significant advantages and beneficial effects compared with prior art, embodies in the following areas：
1. system has the function of that the study of mechanical arm dynamic model and compensation, user can directly be instructed, and not
Additional sensor is needed, without the use of teaching box, reduces production cost.Torsion feedback and motor track by the driver of motor
Instruction can learn the arm dynamic model of mechanical arm, and without using person's input parameter, machinery can be with self-teaching, friendly
Property is good；
2. the mechanical arm dynamic model learnt can permanent memory, therefore there is the mechanical arm learnt to repeat
Study.The dynamic model result for learning can be directly sleeved on identical bar length, material, the mechanical arm of mechanical structure, have logical
With property, accommodation time and time cost can be reduced.Can be arranged in pairs or groups motor encoder according to the mechanical arm dynamic model learnt
The feedback of out position, speed, acceleration is measured, the torque command gone out to calculate each joint motor makes motor export the power
Amount, therefore performed in user and directly instruct process, applied since mechanical arm moves required strength by motor, can
To reach the effect of laborsaving teaching；
3. can automatically or artificial mode is by path optimization, user in controller or can also utilize long-range meter
In the operate interfaces such as calculation machine, compile program pin using path and track is compiled, to ensure the flatness in path and security；
4. the learning dynamics model produced by dynamic learning module, teaching memory module produces the road after mechanical arm study
Footpath, plus path optimization's module optimization, smoothness as the path of the mechanical arm obtained by teaching memory module, is converted to manipulator
The path processing archives of arm, feed path execution module, allowing can make mechanical arm accurately perform the required operation of user
Route.
Brief description of the drawings
Fig. 1：Mechanical arm instructs the schematic diagram of control system；
Fig. 2：The schematic diagram of control device；
Fig. 3：Dynamic learning module forms schematic diagram；
Fig. 4：Memory module is instructed to form schematic diagram；
Fig. 5：Path execution module forms schematic diagram；
Fig. 6：Another structure diagram of control device；
Fig. 7：Path optimization's inside modules component schematic diagram；
Fig. 8：The another structure diagram of control device；
Fig. 9：Instruct the flow diagram of control system.
Embodiment
In order to which the technical features, objects and effects of the present invention are more clearly understood, specific implementation is now described in detail
Scheme.
As shown in Figure 1, mechanical arm teaching control system includes control device 2, operator interface device 3 and mechanical arm 4,
Control device 2 is connected with operator interface device 3, operator interface device 3 can be local terminal PC, long-range PC, industrial computer or
The running gear of any kenel, operator interface device 3 and 2 mode of connection of control device according to operator interface device 3 kenel not
With and difference, such as：Assuming that operator interface device 3 is long-range PC, its connection mode is wireless Internet, is enabled a user to
Control device 2 is convenient to use, mechanical arm 4 includes at least motor driving part 41 and end effector 42.
As shown in Fig. 2, control device 2 includes three modules, be respectively dynamic learning module 21, teaching memory module 22 and
Path execution module 23, but not to be limited.
Fig. 3 forms schematic diagram for dynamic learning module 21, this dynamic learning module 21 includes：Digital independent portion 211, by counting
According to the relevant data setting file 25 of 4 dynamic learning of mechanical arm in 211 read control device 2 of reading part, data setting file
25 can be vector form, bitmap, binary data file or text file all can, content for mechanical arm 4 mechanism length,
Connection mode, kinematic parameter etc.；First order generating unit 212, is established required when learning according to the data in digital independent portion 211
Motor movement track and mechanical arm 4 dynamic model, hereinafter referred to as learn locus model (learning pattern),
The locus model can be the combination of a group formula or a group numeral, also can be the combination of a prescription method；First order output section
213, the locus model generated according to the first order generating unit 212, the motor in mechanical arm 4 is issued commands to according to controller
41 interludes of drive division (follow-up to be known as the interpolation time) calculate the position command of each motor, and position command is defeated
Go out to motor driving part 41；After motor driving part 41 obtains the position command of each motor, i.e., driven according to position command content
Dynamic mechanical arm 4, when 4 start of mechanical arm, can related generation moment of torsion feedback (torque feedback) signals；Dynamic model meter
Calculation portion 215, locus model and 41 processing of the motor driving part formation of the study established out according to the first order generating unit 212
Moment of torsion feedback obtains the learning dynamics model of mechanical arm 4 after calculating, this learning dynamics model contains inertia item, Corrioli's effect
The seven sport physical quantitys such as item, centripetal force item, gravity item, elastic force item, frictional force item, torsion item, but the thing that dynamic model includes
Reason amount is not limited only to this.With following formula 1 for 4 dynamic equation of mechanical arm, it may make up a reality of required locus model during study
Apply example：
Wherein：Q=(q1,q2,...,qn)TFor the set of all motor positions；τ=(τ1,τ2,...,τn)TFor all motors
The set of torsion；M (q) is the inertia item of 4 dynamic of mechanical arm, related with motor position；For the dynamic of mechanical arm 4
Centripetal force and Corrioli's effect item, it is related with motor position with speed；G (q) is the gravity item of 4 dynamic of mechanical arm, with motor position
It is equipped with pass；K (q) is the elastic force of mechanical arm 4, related with motor position；For the frictional force of 4 dynamic of mechanical arm, bag
Power containing Coulomb friction and viscous friction power, it is related with motor speed.The method of the dynamical system study of mechanical arm 4, is to pass through
First order generating unit 212 is formed such as the locus model of formula 1, and the first order output section 213 is by the position command of each motor
It is transmitted to each motor driving part 41；Dynamic model calculating part 215 receives the torque value and first from 41 feedback of motor driving part
The locus model that order generating unit 212 exports can respectively obtain inertia item, Corrioli's effect item, centripetal force item, gravity item, elastic force
Item, frictional force item, and then determine the dynamic model of mechanical arm 4.Dynamic model memory portion 214, by the dynamic of calaculating apparatus arm 4
States model result is stored, this data is used for the dynamic compensating unit 22A for instructing memory module 22.The mechanical arm learnt
4 dynamic model can permanent memory preserve, therefore there is the mechanical arm 4 learnt to be not required to repetitive learning.Dynamic learning module 21
Machine self-teaching is gone out those parameters, this parameter is inputted without using person, the dynamic model result for learning can be applied mechanically directly
In the mechanical arm 4 of same physical material and mechanical structure, there is certain versatility, reduce accommodation time and time cost.
Such as Fig. 4, teaching memory module 22 includes two units, i.e. dynamic compensating unit 22A and mechanism converting unit 22B.
User can utilize operator interface device 3 to open dynamic compensating unit 22A, or via the number in dynamic model memory portion 214
Start teaching memory module 22 according to this.Dynamic compensating unit 22A is included：Dynamic model reading part 221, reads dynamic learning module
The dynamic model data learnt in 21 in dynamic model memory portion 214；Torque command calculating part 222, according to the dynamic of reading
The current motor position coordinate that model data is measured with motor driving part 41 is as motor position signal, to calculate motor
Required torque command (torque command), motor position signal at this time can be considered initial value.Torque command output section
223, the torque command signal that torque command calculating part 222 is formed is transmitted to motor driving part 41, wherein, at motor driving part 41
Reason torque command output section 223 obtains torque command signal, and after driving manipulator arm 4, to generate the horse of different coordinates
Up to position signal, and this motor position signal is returned into torque command calculating part 222 and is handled, mended with forming loop circuit dynamic
Repay.So in cycles, user can terminate this circulation by operator interface device 3, to export final motor position signal.Machine
Structure converting unit 22B is included：Mechanism changes calculating part 225, it measures final motor position using motor driving part 41
The mechanism parameter of the mechanical arm 4 included in signal, with read control device 2 in data setting file 25, and then calculate machine
The location point (forward kinematics) of the end effector (end-effector) 42 of tool arm 4；Path memory portion
224, the location point of the end effector 42 calculated is saved as to the path of mechanical arm 4.The generation of motor movement is usually
User acted caused by directly teaching, allow user be not felt by the inertia force of movable mechanical arm 4, Corrioli's effect,
The active forces such as centripetal force, gravity, frictional force, elastic force, because the strength needed for movable mechanical arm 4 is applied by motor, therefore reach
To the effect of laborsaving teaching, also reach and be not necessary to directly be instructed under additional sensor mode, for the user not only side of operation
Just, cost can also reduce.
Such as Fig. 5, path execution module 23 includes：Path reading part 231, reads the machinery for being stored in path memory portion 224
4 path of arm；Second order generating unit 232, reads related data, this data is included from the data setting file 25 in controller
Kinematic parameter and mechanism parameter and interpolation time (interval time of order), and combine the number of path of above-mentioned mechanical arm 4
According to motor position signal, calculate the position command of each motor；Second order output section 233, generates according to the second order
Portion 232 calculates the position command of each motor, exports to motor driving part 41, after making the execution calculating of end effector 42
Path, formed geometric locus.
The calculating process of the process, i.e. the second order generating unit 232 of the position command described below for forming motor, it is used
The equation of two common inverse kinematics：
xi=Kini(qi, m) and (formula 3)
Wherein：Subscript i represents the value obtained the interpolation time of ith, qiThe motor measured by motor driving part
Position signal；xD, iTo be respectively 42 ideal velocity of current end actuator, position, it is by kinematic parameter and mechanical arm 4
Path and the value that goes out of interpolation temporal calculation；xiFor the physical location of current end actuator 42, can be pushed away by formula 3；KpFor position
Loop gain is put, is a kind of kinematic parameter, from data setting portion 25；Ji(qi, m) and it is Jacobian matrix (Jacobian
Matrix), the calculating of this matrix needs motor position signal (qi) information with the mechanism parameter (m) of mechanical arm 4, its matrix will
Motor coordinate space maps to the location point coordinate space of end effector 42, its image mode is known techniques, in this no longer
Repeat；It is the physical location x of end effector 42 described in formula 3iAcquisition pattern, wherein KiniFunction is mechanism parameter and motor
The function that position signal is formed.The value obtained by the inverse operation from Jacobian matrix, each motor can be learnt by being calculated through formula 2
Amount of movement δ qi, then by this amount of movement summation (Σ) can obtain the position command of motor down between the different poor added times.
Foregoing describe mechanical arm 4 how the function of self-teaching dynamic model and self-compensating, user can save
Power, intuitively directly instructed, and be not required to additional sensor, reduce production cost.The dynamic model result for learning can
Set is used in identical mechanical arm 4, has certain versatility, reduces accommodation time and human cost.
In addition this system can automatically or manually be carried out teaching path optimization, with remove undesirable location point,
Increase path accuracy and flatness, its another embodiment is path optimizing as shown in fig. 6, also include path optimization's module 24
To obtain the path needed for user.Path optimization's module 24 is as shown in fig. 7, path optimization's module 24 includes path reading part
241, read the path for the mechanical arm 4 for being stored in path memory portion 224；Portion of path optimization 242, automatically by original mechanical arm
4 path optimization, or alternative routing, make path smooth to reach the demand of user；This portion of path optimization 242 can more wrap
Automatic Optimal portion (not shown in the figure) and artificial optimization portion (not shown in the figure) are included, Automatic Optimal portion includes one or more
Program, when path Automatic Optimal portion starts, program can start path smooth function and reach Automatic Optimal；Artificial optimization portion can
Program (not shown in the figure) is compiled including path, it can be graphic interface (GUI) or word interface that program is compiled in this path, make
User can utilize the program viewing included in artificial optimization portion this time teaching path and teaching path and context, workpieces processing
Position relative relation.If path is not up to the standard set by user, such as path is not smooth enough, path excessively mechanization
Etc., smoothness or the profile errors of this program setting track can be used with the flatness of adjusts path, or adjustment in user
The number of control point (control point, node) or position carry out adjusts path and avoid external interference in path, to ensure path
Flatness and security, until ideal path.Automatic and artificial optimization portion is using upper unlimited number, also unlimited sequencing,
Or it is used for multiple times and coordinates different order also may be used；Path optimizing memory portion 243, it would be desirable to path record be converted into mechanical arm
4 path processing archives or document feed path execution module 23 use.Processing archives and file are not limiting as form, word
File, bitmap or vectogram etc., the file format that any path execution module 23 can be read can.
Path optimization's module 24 can be used not only in picking and placeing, while can also be used in dispensing, welding, polishing, and path accuracy will
Ask higher processing.Path optimization's module 24 can carry out the amendment of route with computerization or artificial method, particularly teach
There may be the environmental factor more than a variety of not consider in guiding path, therefore cause road through being unable to reach the demand of user, therefore in terms of
Calculation machine or artificial method optimizing are preferably to select.Path optimization's module 24 provides the path of subsequent mechanical arm 4
Modification, makes learning path more accurate.
If Fig. 8 is further embodiment of this invention, similar to the control device 2 of Fig. 6, it includes dynamic learning module 21, religion
Lead memory module 22, path execution module 23 and path optimization's module 24.It is path optimization's module 24 and road with Fig. 6 differences
The execution sequence of footpath execution module 23 is different.User can voluntarily change the order according to use demand, that is, first carry out road
Execution module 23 driving manipulator arm 4 in footpath reuses the optimization operation order of path optimization's module 24 after performing courses of action, or
The path optimizing processed file that path optimization's module 24 generates also may be used for path optimization's module 24.Path execution module 23 with
Path optimization's module 24 also can perform multiple.The order and number that path execution module 23 is performed with path optimization's module 24 do not exist
In limited range of the present invention, the present invention is only its preferred embodiment (two modules are all using once).
Such as Fig. 9, mechanical arm instructs the Optimization Steps operating process of control system.Use path optimization's mould described in Fig. 7
Block 24 carries out.Illustrate flow below：
Step 51：Use dynamical system study module Learning machine arm dynamic model.In this step, by dynamic learning
The first order generating unit 212 in module 21 produces the torsion feedback value of locus model and motor driving part 41 in mechanical arm 4,
Robotic arm dynamic model can be established out；
Step 52：When user directly instructs, teaching logging modle starts, and records the end effector 42 of mechanical arm 4
Position and compensate the dynamic of mechanical arm 4 to reach labor-saving effect.In this step, logging modle is instructed to start, wherein
The encoder position of motor is converted into after the position coordinates of end effector 42 to form path by mechanism converting unit 22B；And
Dynamic compensating unit 22A will calculate each motor automatically under the position, speed, acceleration by the information of motor encoder
Torsion order, make motor outputting torsion bid value, therefore, user directly instruct during without applying power in machinery
On arm 4, to reach labour-saving effect；
Step 53：Path optimization's module optimizes the path of record.It is by recorded in step 52 in this step 53
Mechanical arm 4 end effector 42 path carry out Automatic Optimal, by path go removal of impurities point smoothed；
Step 54：User checks whether the path of Automatic Optimal is preferable, judges whether to compile manually to carry out path.
In this step, user, using path optimization's module 24, is emulated and is performed to watch mechanical arm 4 from operator interface device 3
Path action is instructed, whether viewing path has preferable flatness, whether collided with context generation, interference；
Step 55：User compiles program progress path optimization using path and reaches ideal path.It is basis in this step
Step 54 judges, if path is undesirable, user can use path optimization's module 24 in any operator interface device 3
Program is compiled in path, in a manner of artificial optimization or device Automatic Optimal, is compiled again for automatic smoothing track, directly
Untill the ideal of path；
Step 56：Mechanical arm performs the path processing archives of optimization.After this step is judged according to step 54, if
The path for checking Automatic Optimal is preferable path, needs not move through optimization, execution module 23 control machinery arm 4 in path performs excellent
The teaching path file of change；In addition, step 56 can also be in step 55 by artificial optimization path or device automation path
Afterwards, execution module 23 control machinery arm 4 in path performs the teaching path file after optimization.
System has the function of the study of mechanical arm dynamic model and compensation, and user can directly be instructed, and be not required to
Additional sensor, without the use of teaching box, reduces production cost.Torsion feedback and the finger of motor track by the driver of motor
Order can learn the arm dynamic model of mechanical arm, and without using person's input parameter, machinery can be with self-teaching, friendly property
It is good.The mechanical arm dynamic model learnt can permanent memory, therefore have the mechanical arm learnt that repetitive learning is not required.Institute
The dynamic model result for learning can be directly sleeved on identical bar length, material, the mechanical arm of mechanical structure, have versatility, can
Reduce accommodation time and time cost.Position can be measured according to the mechanical arm dynamic model collocation motor encoder learnt
Put, the feedback of speed, acceleration, the torque command gone out with calculating each joint motor makes motor export the strength, therefore
Performed in user and directly instruct process, applied since mechanical arm moves required strength by motor, can reached
The effect of laborsaving teaching.Can automatically or artificial mode can also be in controller or utilization by path optimization, user
In the operate interfaces such as remote computer, compile program pin using path and track compiled, with ensure the flatness in path with
Security.
Teaching memory module processing, teaching memory module production are given by the learning dynamics Model Transfer that dynamic learning module produces
Path after raw mechanical arm study, and path execution module processing is transferred to, therefore mechanical arm can be made accurately to perform user
Required operation route.
The learning dynamics model produced by dynamic learning module, teaching memory module produce the road after mechanical arm study
Footpath, plus path optimization's module optimization, smoothness as the path of the mechanical arm obtained by teaching memory module, is converted to manipulator
The path processing archives of arm, feed path execution module, allowing can make mechanical arm accurately perform the required operation of user
Route.
In conclusion the present invention can reach the effect of laborsaving teaching, and it is not necessary to installation sensor, allows mechanical arm to learn
To dynamic model.After directly instructing, path optimization's module automatically can also connect path optimization, user in any operation
On mouth, compile program using path and track is compiled, preferable path is finally changed into the processing that robot can be read
Shelves；Last robot performs this processing shelves and reappears the teaching action optimized.Production cost can be not only reduced, can also achieve path
The demand of precision.
It should be noted that：The foregoing is merely the preferred embodiment of the present invention, is not limited to power of the invention
Sharp scope；At the same time more than description, should can understand and implement for the special personage of correlative technology field, thus it is other without departing from
The equivalent change or modification completed under disclosed spirit, should be included in claim.
Claims (10)
1. mechanical arm instructs control system, it is characterised in that：Comprising operator interface device, control device and mechanical arm,
The operator interface device is connected with the control device, and the operator interface device is described to control the control device
Control device is connected with mechanical arm, and the mechanical arm includes an at least motor driving part and an end effector, the horse
Up to drive division to driving manipulator arm；
The control device includes dynamic learning module, teaching memory module and path execution module,
The dynamic learning module, including：Digital independent portion, to the data setting file of read control device；
First order generating unit, study locus model is established according to data setting file；
First order output section, according to study locus model, calculates the position command of each motor and exports to motor and drive
Dynamic portion, wherein motor driving part produce a motor torsional moment feedback according to the position command of each motor；
Dynamic model calculating part, the learning dynamics model of mechanical arm is obtained according to study locus model and motor torsional moment feedback；
Dynamic model memory portion, to store the learning dynamics model of mechanical arm；
The teaching memory module, comprising dynamic compensating unit and mechanism converting unit,
Dynamic compensating unit, including：Dynamic model reading part, to read the mechanical arm being stored in dynamic model memory portion
Learning dynamics model；
Torque command calculating part, learning dynamics model and the motor driving of the mechanical arm according to stored by dynamic model memory portion
The motor position signal that portion is measured forms a torque command signal；
Torque command output section, receive the torque command signal that is formed of torque command calculating part and produce torque command export to
Motor driving part；The torque command signal driving manipulator that wherein motor driving part processing is obtained by torque command output section
Arm, and motor position signal is generated, the processing of torque command calculating part is back to, is compensated with forming the dynamic of a loop circuit；
Mechanism converting unit, including：Mechanism changes calculating part, according to motor position feedback signal and the mechanism parameter of mechanical arm
Calculate the first position point of the end effector of mechanical arm；
The first position point of the end effector of mechanical arm, is stored in the courses of action for mechanical arm by path memory portion；
The path execution module, including：Path reading part, to read the operation for the mechanical arm for being stored in path memory portion
Path；
Second order generating unit, according to performing a perform track curve of mechanical arm and the position of end effector to calculate
The position command of each motor；
Second order output section, the position command of each motor of the second order generating unit generation is exported to motor driving part,
The data that mechanical arm exports the second order output section perform courses of action with driving manipulator arm.
2. mechanical arm according to claim 1 instructs control system, it is characterised in that：The control device also includes road
Footpath optimization module, path optimization's module, including：Path reading part, to read the manipulator for being stored in path memory portion
The courses of action of arm；
Portion of path optimization, at least one miscellaneous point is to optimize courses of action in the courses of action to filtration machinery arm；
Path optimizing memory portion, the courses of action optimized to be converted into the processing shelves of mechanical arm.
3. mechanical arm according to claim 2 instructs control system, it is characterised in that：The portion of path optimization includes one
Automatic Optimal portion and an artificial optimization portion.
4. the mechanical arm teaching control system according to Claims 2 or 3, it is characterised in that：The portion of path optimization root
The courses of action of simulation manipulator arm are carried out according to courses of action and a surrounding environment, a position relative relation of a workpieces processing
Act to judge whether to adjust the operation of mechanical arm.
5. mechanical arm according to claim 3 instructs control system, it is characterised in that：The path artificial optimization portion tune
The operation of whole mechanical arm includes one smooth trajectory degree of setting and/or a profile errors.
6. mechanical arm according to claim 1 or 2 instructs control system, it is characterised in that：The position life of each motor
Order is sent to mechanical arm by dynamic learning module via motor driving part.
7. mechanical arm according to claim 1 or 2 instructs control system, it is characterised in that：Motor torsional moment feedback passes through
Motor driving part is sent to the dynamic model calculating part of dynamic learning module.
8. mechanical arm according to claim 1 instructs control system, it is characterised in that：The control device also includes one
To optimize path optimization's module of mechanical arm courses of action.
9. mechanical arm according to claim 8 instructs control system, it is characterised in that：Path optimization's module bag
Include：Path reading part, to read the courses of action of mechanical arm executed；
The excellent dynamicization portion in path, to filter at least one miscellaneous point of the courses of action of tool arm executed to optimize the behaviour of executed
Make path；
Path optimizing memory portion, the courses of action of the executed optimized to be converted into a processing shelves of mechanical arm.
10. mechanical arm according to claim 1 or 2 instructs control system, it is characterised in that：The moment of torsion life of each motor
Order is sent to mechanical arm by dynamic compensating unit via motor driving part, and motor position signal is transmitted by motor driving part
Compensated to the torque command calculating part of dynamic compensating unit with forming loop circuit dynamic.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711082492.3A CN107972029A (en) | 2017-11-07 | 2017-11-07 | Mechanical arm instructs control system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201711082492.3A CN107972029A (en) | 2017-11-07 | 2017-11-07 | Mechanical arm instructs control system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107972029A true CN107972029A (en) | 2018-05-01 |
Family
ID=62013157
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201711082492.3A Withdrawn CN107972029A (en) | 2017-11-07 | 2017-11-07 | Mechanical arm instructs control system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107972029A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112824060A (en) * | 2019-11-21 | 2021-05-21 | 财团法人工业技术研究院 | Machining route generating device and method |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI594858B (en) * | 2016-12-29 | 2017-08-11 | 新代科技股份有限公司 | Robotic arm teaching system |
-
2017
- 2017-11-07 CN CN201711082492.3A patent/CN107972029A/en not_active Withdrawn
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TWI594858B (en) * | 2016-12-29 | 2017-08-11 | 新代科技股份有限公司 | Robotic arm teaching system |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112824060A (en) * | 2019-11-21 | 2021-05-21 | 财团法人工业技术研究院 | Machining route generating device and method |
| CN112824060B (en) * | 2019-11-21 | 2023-03-28 | 财团法人工业技术研究院 | Machining route generating device and method |
| US11648667B2 (en) | 2019-11-21 | 2023-05-16 | Industrial Technology Research Institute | Processing path generating device and method thereof |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN1903523B (en) | Method of evaluating and correcting robot program and device for evaluating and correcting robot program | |
| US10864632B2 (en) | Direct teaching method of robot | |
| JP6669715B2 (en) | Vibration suppressor | |
| CN111015649B (en) | Driving and controlling integrated control system | |
| CN110053044A (en) | A kind of parallel robot string class fruit clamping model-free adaption Smooth Sliding-Mode impedance adjustment | |
| JP2008207262A (en) | Manipulator system | |
| CN111230882A (en) | Self-adaptive variable impedance control method for fruit sorting parallel robot clamping mechanism | |
| Slawiñski et al. | PD-like controller with impedance for delayed bilateral teleoperation of mobile robots | |
| CN111775145A (en) | Control system of series-parallel robot | |
| TWI594858B (en) | Robotic arm teaching system | |
| CN107972029A (en) | Mechanical arm instructs control system | |
| JP4829151B2 (en) | Robot program evaluation / correction method and robot program evaluation / correction device | |
| EP4241930A1 (en) | Robot control in working space | |
| WO2020149020A1 (en) | Robot control device, robot control method, and robot control program | |
| CN114102636B (en) | Welding seam polishing control system of teleoperation robot and design method and application thereof | |
| CN109664271A (en) | A kind of tele-robotic teaching system based on virtual reality | |
| Fateh et al. | On the transforming of control space by manipulator Jacobian | |
| Nuritdinovich et al. | The concept of the mathematical description of the multi-coordinate mechatronic module of the robot | |
| WO2021241512A1 (en) | Control device, robot system, and control method for causing robot to execute work on workpiece | |
| JPH0635525A (en) | Robot arm control method | |
| CN110488608B (en) | Intelligent kinetic parameter identification method and module for driving and controlling integrated control system | |
| JP2020055095A (en) | Control device and control system | |
| Marsono et al. | Calibration the travel resolution stepper motor machines of CNC-based batik robot running GRBL firmware | |
| Manasra et al. | Design, control and implementation of SCARA robot for sorting missions with machine vision | |
| Kim et al. | A Study on Track Record and Trajectory Control of Articulated Robot Based on Monitoring Simulator for Smart Factory |
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 | ||
| CB02 | Change of applicant information | ||
| CB02 | Change of applicant information |
Address after: 215000 Chunhui Road, Suzhou Industrial Park, Jiangsu Province Applicant after: New Generation Science and Technology (Suzhou) Co., Ltd. Address before: 215021 No. 9 Chunhui Road, Suzhou Industrial Park, Jiangsu Province Applicant before: Suzhou Xindai Numerical Control Equipment Co., Ltd. |
|
| WW01 | Invention patent application withdrawn after publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20180501 |