CN105643625B - A kind of single site based on mechanical arm is fed the method for work of production and processing system - Google Patents
A kind of single site based on mechanical arm is fed the method for work of production and processing system Download PDFInfo
- Publication number
- CN105643625B CN105643625B CN201610128381.0A CN201610128381A CN105643625B CN 105643625 B CN105643625 B CN 105643625B CN 201610128381 A CN201610128381 A CN 201610128381A CN 105643625 B CN105643625 B CN 105643625B
- Authority
- CN
- China
- Prior art keywords
- mrow
- msub
- msup
- mechanical arm
- msubsup
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
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
- B25J9/1656—Programme controls characterised by programming, planning systems for manipulators
-
- 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
Abstract
The invention discloses the working mechanism that a kind of single site based on mechanical arm is fed production and processing system, it is characterized in that single site feed production and processing system includes：Warehouse for finished product and workpiece that caching storehouse that mechanical arm, conveyer belt, industrial camera, capacity are M, capacity are N.Mechanical arm obtains caching the storehouse vacant amount m and handgrip position p under current state, according to an optimal control policy v*To select forward sight distance.If in forward sight distanceInside there is workpiece then to carry out unloading operation, if in forward sight distanceInterior no workpiece then carries out service operations.Wherein optimal control policy v*It is to optimize solution to the SMDP models set up as state to cache storehouse vacant amount m and handgrip position p by Policy iteration algorithm to obtain.The present invention can improve the process balance and production efficiency of the single site mechanical arm production line processed for on-fixed beat, random feed, on-fixed point, and foundation is provided for the single site mechanical arm production line Optimized Operation in industrial production.
Description
Technical field
The present invention relates to automation field, especially a kind of work of single site conveyer belt feed production and processing station system
Method.
Background technology
With the quick development of modern industry, mechanical arm automatic production line is increasingly wide in the application of field of industrial production
It is general, such as in electronic manufacture, automobile making, processing and packing, goods sorting field.Particularly machine vision technique is on a production line
Utilization, drastically increase the flexibility of system, intelligent and automatization level.The generally configuration one of such automatic production line
It is individual or it is multiple for the mechanical arm that picks or process as specific executing agency；Configuring one or more conveyer belt is used to convey work
Part and packing box；Configuring industrial vision system is used for the different operations such as positioning, identification, the dimensional measurement of workpiece.
However as the development of social economy, the particularization of production and processing, specialization, randomization are to automatic production line
Production efficiency, production versatility, intelligent and flexibility propose higher requirement.Traditional production line often adapts to single
Workpiece stream, and process workpiece in fixing point according to fixed beat.For on-fixed beat, random feed, multi-varieties and small-batch
Adaptability and production efficiency etc. the mode of production is not high.
The content of the invention
The present invention is that there is provided a kind of single site feed based on mechanical arm in order to overcome the weak point that prior art is present
The method of work of production and processing system, stands to improve for on-fixed beat, random feed, the single of on-fixed point processing
Point mechanical arm production line production efficiency so that for the single site mechanical arm production line Optimized Operation in industrial production provide according to
According to.
The present invention adopts the following technical scheme that to solve technical problem：
A kind of single site based on mechanical arm of the present invention is fed the method for work of production and processing system, the single site feed
Production and processing system includes：Warehouse for finished product and work that caching storehouse that mechanical arm, conveyer belt, industrial camera, capacity are M, capacity are N
Part；
The mechanical arm is located at the side of the conveyer belt, and the caching storehouse is respectively arranged with the both sides of the mechanical arm
And warehouse for finished product；The industrial camera is in the upstream of the mechanical arm, and vertical just to the workpiece on the conveyer belt, and order is described
Industrial camera it is vertical just to position be the point P that takes picturescam；And the industrial camera can will be located at the point P that takes picturescamThe workpiece position in downstream
Put and pass to the mechanical arm；The preceding viewpoint for defining the mechanical arm is Plook, and positioned at the preceding viewpoint PlookThe workpiece in downstream
It can not be captured by the mechanical arm；
Vacant amount in definition caching storehouse is m；m∈[0,M]；The position for defining the handgrip of the mechanical arm is p；Work as p=1
When, represent the position of handgrip of the mechanical arm in warehouse for finished product；As p=0, the position p of the handgrip of the mechanical arm is represented
In caching storehouse；The united state S of the system of processing is made up of vacant amount m and handgrip position pm,p；Definition is with the forward sight
Point PlookIt is forward sight distance for one section of observed range of starting pointWorkpiece is defined by before described
The apparent distanceThe time spent is the forward sight time
Working range of the mechanical arm on the conveyer belt is defined for Wpick, and in the working range WpickInside enter
Row unloading operation, service operations are carried out on caching storehouse；The unloading operation is to capture the workpiece from the conveyer belt
Into the caching storehouse；The service operations are to be put into after the workpiece in the caching storehouse is processed in the warehouse for finished product；
The time that definition completes unloading operation is discharge timeThe time that definition completes service operations is service timeComplete
The handgrip of the mechanical arm is in the caching storehouse after into unloading operation；The handgrip of the mechanical arm after service operations are completed
In the warehouse for finished product；It is characterized in, the method for work is to carry out as follows：
Step 1, the system of processing bring into operation, and workpiece reaches mechanical arm operation interval at random；Setting line speed,
Mechanical arm speed, workpiece arrival rate, work pieces process rate；Defined variable i, and initialize i=1；
Step 2, the mechanical arm obtain the vacant amount m of ithiWith handgrip position pi, according to optimal control policy v*Selection
Ith forward sight distance
Step 3, the optimal forward sight distance for judging the ithInside whether there is workpiece, if there is workpiece, the machinery
Arm is to the optimal forward sight distance in the ithIt is interior near preceding viewpoint PlookWorkpiece carry out unloading operation after, perform
Step 4；If without workpiece, the mechanical arm is carried out after service operations to the workpiece in the caching storehouse, performs step 5；
Step 4, judge near preceding viewpoint PlookWorkpiece ith discharge timeWhether be more than it is described near
Preceding viewpoint PlookWorkpiece the ith forward sight timeIf being more than, i+1 is assigned to after i, return to step 2 is performed；It is no
Then, the unloading delay by ithAfterwards, i+1 is assigned to i, and return to step 2 is performed；
Step 5, judge near preceding viewpoint PlookWorkpiece ith service timeWhether be more than it is described near
Preceding viewpoint PlookWorkpiece the ith forward sight timeIf being more than, i+1 is assigned to after i, return to step 2 is performed；It is no
Then, service and be delayed by ithAfterwards, i+1 is assigned to i, and return to step 2 is performed.
The characteristics of single site of the present invention based on mechanical arm is fed the method for work of production and processing system lies also in：
The system of processing is to control the forward sight distance using vacant the amount m and handgrip position p as state's
Length, and then control the mode of operation of the mechanical arm；
When vacant amount m is smaller, then the mechanical arm chooses shorter forward sight distanceThe mechanical arm is intended to clothes
Business operation；
When vacant amount m is larger, then the mechanical arm chooses longer forward sight distanceThe mechanical arm is intended to unload
Carry operation.
The optimal control policy v*Obtain according to the following steps：
Step 1, the system of processing are used as state, forward sight distance using the vacant amount and handgrip position for caching storehouseChoosing
It is taken as, to take action, setting up half Markov decision models；
Step 2, solution optimized by double of Markov model of Policy iteration algorithm, obtain optimal control policy v*。
Half Markov model is set up according to the following steps：
Step 1, the state in the definition caching storehouse are its vacant amount m, then the state space in the caching storehouse is Φ1, and
There is Φ1=0,1 ..., M }；
The state space for defining handgrip position p is Φ2, and have Φ2={ 0,1 }；As p=0, the mechanical arm is represented
The position p of handgrip is in caching storehouse, as p=1, represents the position of handgrip of the mechanical arm in warehouse for finished product；
Define the united state S of the system of processingm,pState space be Φ, and have Φ=Φ1×Φ2；
Step 2, the former apparent distanceAs the system of processing in united state SmpUnder action, the system of processing
A steady control strategy v be all united states to the mapping taken action, and haveWhereinIt is m to represent the vacant amount worked as in caching storehouse, and processing is when the position of handgrip is in caching storehouse
The action of system；Represent when the vacant amount in caching storehouse is m, the action of system of processing when the position of handgrip is in warehouse for finished product；It is fixed
Adopted All Policies v collection is combined into Ω, and Ω=v | v=(v (1), v (2), v (3) ..., v (N)), v (i) ∈ D }；The processing system
S is not present in systemM,0And S0,1The two states, then corresponding action is designated as NaN；
Step 3, definition TiFor the ith decision-making moment of the system of processing, T is made0=0；The system of processing is defined to carry out
The unloading operation during processing of ith Markov markov properties, which is delayed, isService operations are delayed
When the ith operation of the mechanical arm is unloading operation, then i+1 time decision-making moment Ti+1For：
I+1 time united state is Xi+1=Sm-1,0, expression handgrip (7) position is the vacant amount in p=0, caching storehouse (4)
For m-1；
Ith unloading operation is delayed：
When the ith operation of the mechanical arm is service operations, then i+1 time decision-making moment Ti+1For：
I+1 time united state is Xi+1=Sm+1,1；Expression handgrip (7) position is the vacant amount in p=1, caching storehouse (4)
For m+1；
Ith service operations are delayed：
Define half Markov cores Qv(t) it is：
In formula (5),The operator that representing matrix is multiplied；T represents system time；PvRepresent that Markov is embedded in chain in tactful v
Under transfer matrix, Fv(t) the sojourn time distribution matrix under tactful v is represented, and is had：
In formula (6)It is from state Sm,pTake actionIt is transferred to Sm',p'Probability, in formula (7)It is in state Sm,pTake actionIt is S to NextStatem',p'Sojourn time distribution；
Step 4, definition expected performance function fvFor：
In formula (8),Represent that system of processing is taken actionFrom united state Sm,pIt is transferred to next
State Sm',p'Unit interval expect cost；
The SMDP models of step 5, foundation as shown in formula (9)：
X=(Xt,Φ,D,Qv(t),fv) (9)
In formula (9), XtRepresent the state procedure of t system of processing.
The Policy iteration algorithm is carried out according to the following steps：
Step 1, according to half Markov cores Qv(t) with expected performance function fv, define Equivalent InfinitesimalWith performance matrix of equal valueAnd by SMDP model X=(Xt,Φ,D,Qv(t),fv) be converted to continuous time Markov decision process
α is discount factor, α ∈ [0,1]；
Step 2, defined variable k, define vkFor the kth time strategy of system of processing；Initialize k=0；And the 0th strategy v0
For：
Step 3, formula (11) is utilized to obtain kth time strategy vkPerformance potential vector
In formula (11)；I is unit diagonal matrix；E is unit column vector；And have：
The average waiting time in state i is represented,Represent the stable state point of kth time in equivalence Markov decision processes
Cloth；And have：
Step 4, formula (15) is utilized to obtain kth+1 strategy vk+1：
Step 5, formula (11) is utilized to obtain kth+1 strategy vk+1Performance potential vector
Step 6, judgementWhether set up, ξ is a less constant；If so, then represent obtained the
K+1 strategy vk+1As optimal control policy v*, and algorithm is exited, otherwise；K+1 is assigned to after k, return to step 4.
Compared with the prior art, beneficial effects of the present invention are embodied in：
The present invention, can using a kind of single site mechanical arm production and processing system method of work based on half Markov model
Effective processing on-fixed beat, the production line processing problems that on-fixed picks point, workpiece is reached at random, improve production line each
The balance and production efficiency of process.
1st, the present invention is modeled using half Markov model to single site mechanical arm production and processing system, random arrival
Workpiece stream be divided into the independent workpiece sequences for meeting Markov process of many segments, compared to conventional method according to fixed beat
Workpiece is handled, the present invention, which can be handled more effectively, random reaches the picking of workpiece, processing problems.
2nd, the present invention before the apparent distance control as system of processing action, can according to the real-time workpiece of production line,
The situation of storehouse and warehouse for finished product is cached, Processing Strategies are adjusted.The balance between production line each operation is effectively increased, life is strengthened
The robustness of producing line.
3rd, the present invention optimizes solution using Policy iteration algorithm to the SMDP models of system of processing, compared to other calculations
Method, Policy iteration algorithm avoids local optimum, and fast convergence rate can reduce system of processing and solve the time, improve production
Efficiency.
4th, mechanical arm of the present invention picks mode using removable, is calculated according to the location of workpiece on operation interval and picks experience
Point, can reduce the time that uninstall process waits workpiece.
Brief description of the drawings
Fig. 1 is single site of the present invention feed production and processing system schematic；
Fig. 2 is the flow chart of work methods that single site of the present invention is fed production and processing system.
Embodiment
Referring to Fig. 1, single site feed production and processing system includes in the present embodiment：Mechanical arm 1, conveyer belt 2, industrial camera
3rd, warehouse for finished product 5 and workpiece 6 that capacity is M caching storehouse 4, capacity are N；
Mechanical arm 1 is located at the side of conveyer belt 2, and mechanical arm is 6DOF series connection mechanical arm, in the ring flange of mechanical arm 1
It is upper that pneumatic gripping device 7 is housed.It is V to define mechanical arm speedrobot, in the implementation case, Vrobot=20cm/s.The two of mechanical arm 1
Side is respectively arranged with the warehouse for finished product 5 that the caching storehouse 4 that capacity is M and capacity are N；In the implementation case, M=4, N=
2147483648.Industrial camera 3 is in the upstream of mechanical arm 1, and vertical just to the workpiece 6 on conveyer belt 2；Conveyer belt is by servo
Motor drives, and speed can be with accurate adjustment.It is V to define conveyer belt transporting velocitycon, in the implementation case, Vcon=10cm/s；It is fixed
The working range of adopted mechanical arm 1 on conveyor belt 2 is Wpick, wherein WpickLeft end point isRight endpoint isDefine work
Taking pictures for industry camera 3 is a little Pcam；Positioned at the point P that takes picturescamThe information such as the location of workpiece in downstream can pass to mechanical arm 1；Definition
The preceding viewpoint of the mechanical arm 1 is Plook, and positioned at preceding viewpoint downstream PlookWorkpiece will lose；
Vacant amount in definition caching storehouse 4 is m, m ∈ [0,4]；The position for defining the handgrip 7 of mechanical arm 1 is p；Work as p=1
When, represent the position of handgrip 7 of mechanical arm 1 in warehouse for finished product 5；As p=0, represent the position p of handgrip 7 of mechanical arm 1 slow
In warehousing 4；It is P to define mechanical arm handgrip coordinatep；The united state of system of processing is made up of vacant amount m and handgrip 7 position p
Sm,p；Viewpoint P before definitionlookIt is forward sight distance for one section of observed range of starting pointDefinition
Workpiece 6 passes through forward sight distanceThe time spent is the forward sight timeAnd have：
lmax=Pcam-Plook (2)
Mechanical arm is in working range WpickInterior carry out unloading operation, service operations are carried out on caching storehouse 4；Unloading operation is
Workpiece 6 is captured from conveyer belt 2 into caching storehouse 4；Service operations are put into after the workpiece 6 cached in storehouse 4 is processed
In warehouse for finished product 5；The time that definition completes unloading operation is discharge timeWhen definition completes the time of service operations for service
BetweenIt is in the handgrip 7 of mechanical arm 1 after completing unloading operation in caching storehouse 4；Mechanical arm 1 grabs after service operations are completed
Hand 7 is in warehouse for finished product 5；
System of processing controls forward sight distance using vacant amount m and the position p of handgrip 7 as stateLength and machinery
The mode of operation of arm 1；When vacant amount m is smaller, show cache storehouse 4 storage pressure it is smaller, then mechanical arm 1 choose it is shorter before
The apparent distanceMechanical arm 1 is intended to service operations, to reduce the utilization rate that the discharge pressure of conveyer belt 2 improves caching storehouse 4；When
When vacant amount m is larger, show that the storage pressure for caching storehouse 4 is larger, then mechanical arm 1 chooses longer forward sight distanceMechanical arm
1 is intended to unloading operation, to mitigate the storage pressure in caching storehouse 4.
The system of processing is used as state, forward sight distance using the vacant amount and handgrip position 7 for caching storehouse 4Selection make
For action, half Markov decision models are set up according to the following steps：
Step 1, the state in definition caching storehouse 4 are its vacant amount m, then the state space for caching storehouse 4 is Φ1, and have Φ1=
{0,1,...,M}；
The state space for defining the position p of handgrip 7 is Φ2, and have Φ2={ 0,1 }, as p=1, represents grabbing for mechanical arm 1
The position of hand 7 is in warehouse for finished product 5；As p=0, represent the position p of handgrip 7 of mechanical arm 1 in caching storehouse 4；
Define the united state S of system of processingm,pState space be Φ, and have Φ=Φ1×Φ2；
Step 2, the former apparent distanceAs system of processing in united state Sm,pUnder action, the one of the system of processing
Individual steady control strategy v is mapping of all united states to action, and is had：
In formula (4)Represent when the vacant amount in caching storehouse 4 is m, the action of system of processing during the position p=0 of handgrip 7；Represent when the vacant amount in caching storehouse 4 is m, the action of system of processing during the position p=1 of handgrip 7；Define All Policies v's
Collection is combined into Ω；Ω=v | v=(v (1), v (2), v (3) ..., v (N)), v (i) ∈ D }.S is not present in system of processingM,0And S0,1This
Two states, then corresponding action is designated as NaN；
Step 3, definition TiMoment, wherein T are shifted for the system of processing ith0=0.The system of processing is defined to carry out
Unloading operation delay is during the processing of Markov markov propertiesService operations are delayed
When mechanical arm ith operation is unloading operation, then next decision-making moment Ti+1For：
NextState is Xi+1=Sm-1,0；System uninstallation delay time is：
When mechanical arm ith operation is service operations, then next decision-making moment Ti+1For：
NextState is Xi+1=Sm-1,1；System service delay time is：
Define half Markov cores Qv(t) it is：
In formula (9),The operator that representing matrix is multiplied；PvTransfer matrix of the Markov insertion chains under tactful v is represented,
Fv(t) the sojourn time distribution matrix under tactful v is represented, and is had：
In formula (10)It is from state Sm,pTake actionIt is transferred to Sm',p'Probability, and have：
In formula (11)It is in state Sm,pTake actionIt is S to NextStatem',p'Residence time
Distribution.
Step 4, definition expected performance function fvFor：
In formula (13),Represent that system of processing is taken actionFrom united state Sm,pIt is transferred to next
State Sm',p'Unit interval expect cost；
Step 5, set up SMDP models as shown in Equation 14：
X=(Xt,Φ,D,Qv(t),fv) (14)
In formula (14), XtRepresent the state procedure of t.
Solution is optimized to half Markov model according to following steps by Policy iteration algorithm, optimal control is obtained
Make strategy v*, and by optimal control policy v*It is cured in the mechanical arm 1.
Step 1, according to half Markov cores Qv(t) with expected performance function fv, define Equivalent InfinitesimalWith performance of equal value
MatrixAnd by SMDP model X=(Xt,Φ,D,Qv(t),fv) be converted to continuous time Markov decision processWherein α is discount factor, α ∈ [0,1].
Step 2, defined variable k, define vkFor the kth time strategy of system of processing；Initialize k=0；And the 0th strategy v0
For：
Step 3, formula (16) is utilized to obtain kth time strategy vkPerformance potential vector
In formula (16)；I is unit diagonal matrix；E is unit column vector；And have：
The average waiting time in state i is represented,Represent the stable state point of kth time in equivalence Markov decision processes
Cloth；And have：
Step 4, formula (20) is utilized to obtain kth+1 strategy vk+1：
Step 5, formula (16) is utilized to obtain kth+1 strategy vk+1Performance potential vector
Step 6, judgementWhether set up, ξ is a less constant；If so, then represent obtained the
K+1 strategy vk+1As optimal control policy v*, and algorithm is exited, otherwise；K+1 is assigned to after k, return to step 4.
By resulting optimal control policy v*After being cured in the mechanical arm 1, the method for work of system of processing is by such as
Lower step is carried out：
Step 1, the system of processing bring into operation, setting line speed Vcon=10cm/s, mechanical arm speed Vrobot=
20cm/s, workpiece arrival rate λ=1, work pieces process rate μ=2；Defined variable i, and initialize i=1.
Step 2, the mechanical arm 1 obtain the vacant amount m of ithiWith the position p of handgrip 7i, according to optimal control policy v*Choosing
Select ith forward sight distance
Step 3, the optimal forward sight distance for judging the ithInside whether there is workpiece, if there is workpiece, the machinery
1 pair of arm is in the forward sight distance of the ithIt is interior near preceding viewpoint PlookWorkpiece carry out unloading operation after, perform step
Rapid 4；If without workpiece, the workpiece in 1 pair of the mechanical arm caching storehouse 4 is carried out after service operations；Perform step 5；
Wherein uninstall process, which is divided into, picks, places and waits three processes；Select forward sight distance interiorFrom preceding viewpoint Plook
A nearest workpiece is as target workpiece, and definition observes that target workpiece coordinate is Pwork, handgrip is from initial position PpSet out, with
The collision point coordinate definition that track picks workpiece isDefinition movement picks the timeMobile standing time isAnd wait
Time isDefine viewpoint P before target workpiece is reachedlookTime be arrival timeThen：
DefinitionFor the separation of workpiece arrival time, and have：
IfThen directly pick workpiece；Then collision pointThe time is picked with movementMeet equation group：
Stand-by period is：
IfThen mechanical arm handgrip is first moved to operation interval left end pointWait workpiece to reach to carry out again
Pick, then the stand-by period be：
It is mobile to pick the time and be
Handgrip pick after workpiece return caching storehouse 4 placed, move standing time be：
So the time of ith unloading operationFor：
Then next decision-making moment is：
NextState is Xi+1=Sm-1,0；The system delay time is：
Service operations are divided into mobile taking-up workpiece and process, and it is caching storehouse 4 and warehouse for finished product 5 to be equally divided into initial position
Two kinds of situations；If p=0, i.e., initial position is caching storehouse 4, then mechanical arm, which is directly postponed, takes out a workpiece in warehousing 4 and added
Workpiece, is positioned in warehouse for finished product 5 by work after machining；If p=1, i.e., initial position is warehouse for finished product 5, then mechanical arm is first moved
Workpiece is taken out to caching storehouse 4 to be processed again；Define ith service operations time beWherein process time obeys solid
Determine random distribution, its time isDefinition spends the time to be T from warehouse for finished product to caching storehousebank_buffer, then
Then next decision-making moment is：
Then next state is Xi+1=Sm+1,1, the system delay time is：
Step 4, judge near preceding viewpoint PlookWorkpiece ith discharge timeWhether be more than it is described near
Preceding viewpoint PlookWorkpiece the ith forward sight timeIf being more than, i+1 is assigned to after i, return to step 2 is performed；It is no
Then, the unloading delay by ithAfterwards, i+1 is assigned to i, and return to step 2 is performed；
Step 5, judge near preceding viewpoint PlookWorkpiece ith service timeWhether be more than it is described near
Preceding viewpoint PlookWorkpiece the ith forward sight timeIf being more than, i+1 is assigned to after i, return to step 2 is performed；It is no
Then, service and be delayed by ithAfterwards, i+1 is assigned to i, and return to step 2 is performed.
Claims (2)
1. a kind of single site based on mechanical arm is fed the method for work of production and processing system, the single site feed production and processing
System includes：Mechanical arm (1), conveyer belt (2), industrial camera (3), the caching storehouse (4) that capacity is M, the warehouse for finished product that capacity is N
And workpiece (6) (5)；
The mechanical arm (1) is located at the side of the conveyer belt (2), is respectively arranged with the both sides of the mechanical arm (1) described
Cache storehouse (4) and warehouse for finished product (5)；The industrial camera (3) is in the upstream of the mechanical arm (1), and vertical just to the biography
Send the workpiece (6) on band (2), make the industrial camera (3) it is vertical just to position be the point P that takes picturescam；And the industrial camera
(3) the point P that takes pictures will can be located atcamThe location of workpiece in downstream passes to the mechanical arm (1)；Define the forward sight of the mechanical arm (1)
Point is Plook, and positioned at the preceding viewpoint PlookThe workpiece in downstream can not be captured by the mechanical arm (1)；
Vacant amount in definition caching storehouse (4) is m；m∈[0,M]；The position for defining the handgrip (7) of the mechanical arm (1) is p；
As p=1, represent the position of handgrip (7) of the mechanical arm (1) in warehouse for finished product (5)；As p=0, the machinery is represented
The position p of the handgrip (7) of arm (1) is in caching storehouse (4)；The system of processing is constituted by vacant amount m and handgrip (7) position p
United state Sm,p；Definition is with the preceding viewpoint PlookIt is forward sight distance for one section of observed range of starting point Define workpiece (6) and pass through the forward sight distanceThe time spent is the forward sight time
Working range of the mechanical arm (1) on the conveyer belt (2) is defined for Wpick, and in the working range WpickIt is interior
Unloading operation is carried out, service operations are carried out on caching storehouse (4)；The unloading operation is from the transmission by the workpiece (6)
Captured on band (2) into the caching storehouse (4)；The service operations are to be processed the workpiece (6) in the caching storehouse (4)
After be put into the warehouse for finished product (5)；The time that definition completes unloading operation is discharge timeDefinition completes service operations
Time is service timeIt is in the handgrip (7) of the mechanical arm (1) after completing unloading operation in the caching storehouse (4)；
It is in the handgrip (7) of the mechanical arm (1) after completing service operations in the warehouse for finished product (5)；
The system of processing is to control the forward sight distance using vacant amount m and handgrip (7) the position p as state's
Length, and then control the mode of operation of the mechanical arm (1)；
When vacant amount m is smaller, then the mechanical arm (1) chooses shorter forward sight distanceThe mechanical arm (1) is intended to
Service operations；
When vacant amount m is larger, then the mechanical arm (1) chooses longer forward sight distanceThe mechanical arm (1) is intended to
Unloading operation；It is characterized in that, the method for work is to carry out as follows：
Step 1, the system of processing bring into operation, and workpiece reaches mechanical arm operation interval at random；Set line speed, machinery
Arm speed, workpiece arrival rate, work pieces process rate；Defined variable i, and initialize i=1；
Step 2, the mechanical arm (1) obtain the vacant amount m of ithiWith handgrip (7) position pi, according to optimal control policy v*Choosing
Select ith forward sight distanceThe optimal control policy v*Obtain according to the following steps：
Step 2.1, the system of processing are used as state, forward sight distance using the vacant amount and handgrip (7) position for caching storehouse (4)
Selection as action, set up half Markov decision models；Half Markov model is set up according to the following steps：
Step 2.1.1, the state of the definition caching storehouse (4) are its vacant amount m, then the state space of the caching storehouse (4) is
Φ1, and have Φ1=0,1 ..., M }；
The state space for defining handgrip (7) position p is Φ2, and have Φ2={ 0,1 }；As p=0, the mechanical arm (1) is represented
Handgrip (7) position p in caching storehouse (4), as p=1, represent the position of handgrip (7) of the mechanical arm (1) in finished product
In storehouse (5)；
Define the united state S of the system of processingm,pState space be Φ, and have Φ=Φ1×Φ2；
Step 2.1.2, the former apparent distanceAs the system of processing in united state Sm,pUnder action, the system of processing
A steady control strategy v be all united states to the mapping taken action, and haveWherein Represent that the position of handgrip (7) is in caching storehouse (4) when the vacant amount in caching storehouse (4) is m
When system of processing action；Represent that when caching the vacant amount in storehouse (4) be m, when the position of handgrip (7) is in warehouse for finished product (5)
The action of system of processing；The collection for defining All Policies v is combined into Ω, and Ω=v | v=(v (1), v (2), v (3) ..., v (N)), v
(i)∈D}；S is not present in the system of processingM,0And S0,1The two states, then corresponding action is designated as NaN；
Step 2.1.3, definition TiFor the ith decision-making moment of the system of processing, T is made0=0；The system of processing is defined to carry out
The unloading operation during processing of ith Markov markov properties, which is delayed, isService operations are delayed
When the ith operation of the mechanical arm (1) is unloading operation, then i+1 time decision-making moment Ti+1For：
<mrow>
<msub>
<mi>T</mi>
<mrow>
<mi>i</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
</msub>
<mo>=</mo>
<msub>
<mi>T</mi>
<mi>i</mi>
</msub>
<mo>+</mo>
<mi>m</mi>
<mi>a</mi>
<mi>x</mi>
<mo>{</mo>
<msub>
<mi>t</mi>
<msub>
<mi>S</mi>
<mrow>
<msub>
<mi>m</mi>
<mi>i</mi>
</msub>
<mo>,</mo>
<msub>
<mi>p</mi>
<mi>i</mi>
</msub>
</mrow>
</msub>
</msub>
<mo>,</mo>
<msubsup>
<mi>t</mi>
<mrow>
<msub>
<mi>m</mi>
<mi>i</mi>
</msub>
<mo>,</mo>
<msub>
<mi>p</mi>
<mi>i</mi>
</msub>
</mrow>
<mrow>
<mi>l</mi>
<mi>o</mi>
<mi>a</mi>
<mi>d</mi>
</mrow>
</msubsup>
<mo>}</mo>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>1</mn>
<mo>)</mo>
</mrow>
</mrow>
I+1 time united state is Xi+1=Sm-1,0, expression handgrip (7) position is that the vacant amount in p=0, caching storehouse (4) is m-
1；
Ith unloading operation is delayed：
When the ith operation of the mechanical arm (1) is service operations, then i+1 time decision-making moment Ti+1For：
<mrow>
<msub>
<mi>T</mi>
<mrow>
<mi>i</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
</msub>
<mo>=</mo>
<msub>
<mi>T</mi>
<mi>i</mi>
</msub>
<mo>+</mo>
<mi>m</mi>
<mi>a</mi>
<mi>x</mi>
<mo>{</mo>
<msub>
<mi>t</mi>
<msub>
<mi>S</mi>
<mrow>
<msub>
<mi>m</mi>
<mi>i</mi>
</msub>
<mo>,</mo>
<msub>
<mi>p</mi>
<mi>i</mi>
</msub>
</mrow>
</msub>
</msub>
<mo>,</mo>
<msubsup>
<mi>t</mi>
<mrow>
<msub>
<mi>m</mi>
<mi>i</mi>
</msub>
<mo>,</mo>
<msub>
<mi>p</mi>
<mi>i</mi>
</msub>
</mrow>
<mrow>
<mi>s</mi>
<mi>e</mi>
<mi>r</mi>
</mrow>
</msubsup>
<mo>}</mo>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>3</mn>
<mo>)</mo>
</mrow>
</mrow>
I+1 time united state is Xi+1=Sm+1,1；Expression handgrip (7) position is that the vacant amount in p=1, caching storehouse (4) is m+
1；
Ith service operations are delayed：
Define half Markov cores Qv(t) it is：
<mrow>
<msup>
<mi>Q</mi>
<mi>v</mi>
</msup>
<mrow>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<msub>
<mrow>
<mo>&lsqb;</mo>
<mi>Q</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>s</mi>
<mrow>
<mi>m</mi>
<mo>,</mo>
<mi>p</mi>
</mrow>
</msub>
<mo>,</mo>
<msub>
<mi>v</mi>
<msub>
<mi>s</mi>
<mrow>
<mi>m</mi>
<mo>,</mo>
<mi>p</mi>
</mrow>
</msub>
</msub>
<mo>,</mo>
<msub>
<mi>s</mi>
<mrow>
<msup>
<mi>m</mi>
<mo>&prime;</mo>
</msup>
<mo>,</mo>
<msup>
<mi>p</mi>
<mo>&prime;</mo>
</msup>
</mrow>
</msub>
<mo>,</mo>
<mi>t</mi>
<mo>)</mo>
</mrow>
<mo>&rsqb;</mo>
</mrow>
<mrow>
<msub>
<mi>s</mi>
<mrow>
<mi>m</mi>
<mo>,</mo>
<mi>p</mi>
</mrow>
</msub>
<mo>,</mo>
<msub>
<mi>s</mi>
<mrow>
<msup>
<mi>m</mi>
<mo>&prime;</mo>
</msup>
<mo>,</mo>
<msup>
<mi>p</mi>
<mo>&prime;</mo>
</msup>
</mrow>
</msub>
<mo>&Element;</mo>
<mi>&Phi;</mi>
</mrow>
</msub>
<mo>=</mo>
<msup>
<mi>P</mi>
<mi>v</mi>
</msup>
<mo>&CircleTimes;</mo>
<msup>
<mi>F</mi>
<mi>v</mi>
</msup>
<mrow>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
</mrow>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>5</mn>
<mo>)</mo>
</mrow>
</mrow>
In formula (5),The operator that representing matrix is multiplied；T represents system time；PvRepresent that Markov is embedded in chain under tactful v
Transfer matrix, Fv(t) the sojourn time distribution matrix under tactful v is represented, and is had：
<mrow>
<msup>
<mi>P</mi>
<mi>v</mi>
</msup>
<mo>=</mo>
<msub>
<mrow>
<mo>&lsqb;</mo>
<msub>
<mi>p</mi>
<mrow>
<msub>
<mi>s</mi>
<mrow>
<mi>m</mi>
<mo>,</mo>
<mi>p</mi>
</mrow>
</msub>
<msub>
<mi>s</mi>
<mrow>
<msup>
<mi>m</mi>
<mo>&prime;</mo>
</msup>
<mo>,</mo>
<msup>
<mi>p</mi>
<mo>&prime;</mo>
</msup>
</mrow>
</msub>
</mrow>
</msub>
<mrow>
<mo>(</mo>
<msub>
<mi>v</mi>
<msub>
<mi>s</mi>
<mrow>
<mi>m</mi>
<mo>,</mo>
<mi>p</mi>
</mrow>
</msub>
</msub>
<mo>)</mo>
</mrow>
<mo>&rsqb;</mo>
</mrow>
<mrow>
<msub>
<mi>s</mi>
<mrow>
<mi>m</mi>
<mo>,</mo>
<mi>p</mi>
</mrow>
</msub>
<msub>
<mi>s</mi>
<mrow>
<msup>
<mi>m</mi>
<mo>&prime;</mo>
</msup>
<mo>,</mo>
<msup>
<mi>p</mi>
<mo>&prime;</mo>
</msup>
</mrow>
</msub>
<mo>&Element;</mo>
<mi>&Phi;</mi>
</mrow>
</msub>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>6</mn>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<msup>
<mi>F</mi>
<mi>v</mi>
</msup>
<mrow>
<mo>(</mo>
<mi>t</mi>
<mo>)</mo>
</mrow>
<mo>=</mo>
<msub>
<mrow>
<mo>&lsqb;</mo>
<msub>
<mi>F</mi>
<mrow>
<msub>
<mi>s</mi>
<mrow>
<mi>m</mi>
<mo>,</mo>
<mi>p</mi>
</mrow>
</msub>
<msub>
<mi>s</mi>
<mrow>
<msup>
<mi>m</mi>
<mo>&prime;</mo>
</msup>
<mo>,</mo>
<msup>
<mi>p</mi>
<mo>&prime;</mo>
</msup>
</mrow>
</msub>
</mrow>
</msub>
<mrow>
<mo>(</mo>
<mi>t</mi>
<mo>,</mo>
<msub>
<mi>v</mi>
<msub>
<mi>s</mi>
<mrow>
<mi>m</mi>
<mo>,</mo>
<mi>p</mi>
</mrow>
</msub>
</msub>
<mo>)</mo>
</mrow>
<mo>&rsqb;</mo>
</mrow>
<mrow>
<msub>
<mi>s</mi>
<mrow>
<mi>m</mi>
<mo>,</mo>
<mi>p</mi>
</mrow>
</msub>
<msub>
<mi>s</mi>
<mrow>
<msup>
<mi>m</mi>
<mo>&prime;</mo>
</msup>
<mo>,</mo>
<msup>
<mi>p</mi>
<mo>&prime;</mo>
</msup>
</mrow>
</msub>
<mo>&Element;</mo>
<mi>&Phi;</mi>
</mrow>
</msub>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>7</mn>
<mo>)</mo>
</mrow>
</mrow>
2
In formula (6)It is from state Sm,pTake actionIt is transferred to Sm',p'Probability, in formula (7)It is in state Sm,pTake actionIt is S to NextStatem',p'Sojourn time distribution；
Step 2.1.4, definition expected performance function fvFor：
<mrow>
<msup>
<mi>f</mi>
<mi>v</mi>
</msup>
<mo>=</mo>
<msub>
<mrow>
<mo>&lsqb;</mo>
<mi>f</mi>
<mrow>
<mo>(</mo>
<msub>
<mi>s</mi>
<mrow>
<mi>m</mi>
<mo>,</mo>
<mi>p</mi>
</mrow>
</msub>
<mo>,</mo>
<msub>
<mi>v</mi>
<msub>
<mi>s</mi>
<mrow>
<mi>m</mi>
<mo>,</mo>
<mi>p</mi>
</mrow>
</msub>
</msub>
<mo>,</mo>
<msub>
<mi>s</mi>
<mrow>
<msup>
<mi>m</mi>
<mo>&prime;</mo>
</msup>
<mo>,</mo>
<msup>
<mi>p</mi>
<mo>&prime;</mo>
</msup>
</mrow>
</msub>
<mo>)</mo>
</mrow>
<mo>&rsqb;</mo>
</mrow>
<mrow>
<msub>
<mi>s</mi>
<mrow>
<mi>m</mi>
<mo>,</mo>
<mi>p</mi>
</mrow>
</msub>
<mo>,</mo>
<msub>
<mi>s</mi>
<mrow>
<msup>
<mi>m</mi>
<mo>&prime;</mo>
</msup>
<mo>,</mo>
<msup>
<mi>p</mi>
<mo>&prime;</mo>
</msup>
</mrow>
</msub>
<mo>&Element;</mo>
<mi>&Phi;</mi>
</mrow>
</msub>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>8</mn>
<mo>)</mo>
</mrow>
</mrow>
In formula (8),Represent that system of processing is taken actionFrom united state Sm,pIt is transferred to NextState
Sm',p'Unit interval expect cost；
The SMDP models of step 2.1.5, foundation as shown in formula (9)：
X=(Xt,Φ,D,Qv(t),fv) (9)
In formula (9), XtRepresent the state procedure of t system of processing；
Step 2.2, solution optimized by double of Markov model of Policy iteration algorithm, obtain optimal control policy v*；
Step 3, the optimal forward sight distance for judging the ithInside whether there is workpiece, if there is workpiece, the mechanical arm (1)
To the optimal forward sight distance in the ithIt is interior near preceding viewpoint PlookWorkpiece carry out unloading operation after, perform step
Rapid 4；If without workpiece, the mechanical arm (1) is carried out after service operations to the workpiece in the caching storehouse (4), performs step
5；
Step 4, judge near preceding viewpoint PlookWorkpiece ith discharge timeWhether it is more than described near forward sight
Point PlookWorkpiece the ith forward sight timeIf being more than, i+1 is assigned to after i, return to step 2 is performed；Otherwise, pass through
Cross the unloading delay of ithAfterwards, i+1 is assigned to i, and return to step 2 is performed；
Step 5, judge near preceding viewpoint PlookWorkpiece ith service timeWhether it is more than described near forward sight
Point PlookWorkpiece the ith forward sight timeIf being more than, i+1 is assigned to after i, return to step 2 is performed；Otherwise, pass through
Cross ith service delayAfterwards, i+1 is assigned to i, and return to step 2 is performed.
2. the single site according to claim 1 based on mechanical arm is fed the method for work of production and processing system, its feature
It is：
The Policy iteration algorithm is carried out according to the following steps：
Step 1, according to half Markov cores Qv(t) with expected performance function fv, define Equivalent InfinitesimalWith performance matrix of equal valueAnd by SMDP model X=(Xt,Φ,D,Qv(t),fv) be converted to continuous time Markov decision processα is discount factor, α ∈ [0,1]；
Step 2, defined variable k, define vkFor the kth time strategy of system of processing；Initialize k=0；And the 0th strategy v0For：
Step 3, formula (11) is utilized to obtain kth time strategy vkPerformance potential vector
<mrow>
<mo>(</mo>
<mi>&alpha;</mi>
<mi>I</mi>
<mo>-</mo>
<msubsup>
<mi>A</mi>
<mi>&alpha;</mi>
<msub>
<mi>v</mi>
<mi>k</mi>
</msub>
</msubsup>
<mo>+</mo>
<msub>
<mi>&rho;</mi>
<mi>&alpha;</mi>
</msub>
<msubsup>
<mi>e&pi;</mi>
<mi>&alpha;</mi>
<msub>
<mi>v</mi>
<mi>k</mi>
</msub>
</msubsup>
<mo>)</mo>
<msubsup>
<mi>g</mi>
<mi>&alpha;</mi>
<msub>
<mi>v</mi>
<mi>k</mi>
</msub>
</msubsup>
<mo>=</mo>
<msubsup>
<mi>f</mi>
<mi>&alpha;</mi>
<msub>
<mi>v</mi>
<mi>k</mi>
</msub>
</msubsup>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>11</mn>
<mo>)</mo>
</mrow>
</mrow>
In formula (11)；I is unit diagonal matrix；E is unit column vector；And have：
<mrow>
<msubsup>
<mi>h</mi>
<mi>&alpha;</mi>
<msub>
<mi>v</mi>
<mi>k</mi>
</msub>
</msubsup>
<mo>=</mo>
<mrow>
<mo>(</mo>
<mi>I</mi>
<mo>-</mo>
<msubsup>
<mi>Q</mi>
<mi>&alpha;</mi>
<msub>
<mi>v</mi>
<mi>k</mi>
</msub>
</msubsup>
<mo>)</mo>
</mrow>
<mi>e</mi>
<mo>/</mo>
<mi>&alpha;</mi>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>12</mn>
<mo>)</mo>
</mrow>
</mrow>
<mrow>
<msub>
<mi>&rho;</mi>
<mi>&alpha;</mi>
</msub>
<mo>=</mo>
<mi>m</mi>
<mi>a</mi>
<mi>x</mi>
<mo>{</mo>
<msubsup>
<mi>h</mi>
<mi>&alpha;</mi>
<msub>
<mi>v</mi>
<mi>k</mi>
</msub>
</msubsup>
<msup>
<mrow>
<mo>(</mo>
<mi>i</mi>
<mo>)</mo>
</mrow>
<mrow>
<mo>-</mo>
<mn>1</mn>
</mrow>
</msup>
<mo>,</mo>
<mi>i</mi>
<mo>=</mo>
<mn>1</mn>
<mo>,</mo>
<mn>2...</mn>
<mi>N</mi>
<mo>,</mo>
<msub>
<mi>v</mi>
<mi>k</mi>
</msub>
<mo>&Element;</mo>
<mi>&Omega;</mi>
<mo>,</mo>
<mi>&alpha;</mi>
<mo>&Element;</mo>
<mo>&lsqb;</mo>
<mn>0</mn>
<mo>,</mo>
<mn>1</mn>
<mo>&rsqb;</mo>
<mo>}</mo>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>13</mn>
<mo>)</mo>
</mrow>
</mrow>
The average waiting time in state i is represented,Represent the steady-state distribution of kth time in equivalence Markov decision processes；
And have：
<mrow>
<mtable>
<mtr>
<mtd>
<mrow>
<msubsup>
<mi>A</mi>
<mi>&alpha;</mi>
<msub>
<mi>v</mi>
<mi>k</mi>
</msub>
</msubsup>
<msubsup>
<mi>&pi;</mi>
<mi>&alpha;</mi>
<msub>
<mi>v</mi>
<mi>k</mi>
</msub>
</msubsup>
<mo>=</mo>
<mn>0</mn>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mrow>
<msubsup>
<mi>&pi;</mi>
<mi>&alpha;</mi>
<msub>
<mi>v</mi>
<mi>k</mi>
</msub>
</msubsup>
<mi>e</mi>
<mo>=</mo>
<mn>1</mn>
</mrow>
</mtd>
</mtr>
<mtr>
<mtd>
<mrow>
<msubsup>
<mi>A</mi>
<mi>&alpha;</mi>
<msub>
<mi>v</mi>
<mi>k</mi>
</msub>
</msubsup>
<mi>e</mi>
<mo>=</mo>
<mn>0</mn>
</mrow>
</mtd>
</mtr>
</mtable>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>14</mn>
<mo>)</mo>
</mrow>
</mrow>
Step 4, formula (15) is utilized to obtain kth+1 strategy vk+1：
<mrow>
<msub>
<mi>v</mi>
<mrow>
<mi>k</mi>
<mo>+</mo>
<mn>1</mn>
</mrow>
</msub>
<mo>&Element;</mo>
<mi>arg</mi>
<munder>
<mi>min</mi>
<mrow>
<mi>v</mi>
<mo>&Element;</mo>
<mi>&Omega;</mi>
</mrow>
</munder>
<mo>{</mo>
<msubsup>
<mi>A</mi>
<mi>&alpha;</mi>
<mi>v</mi>
</msubsup>
<msubsup>
<mi>g</mi>
<mi>&alpha;</mi>
<msub>
<mi>v</mi>
<mi>k</mi>
</msub>
</msubsup>
<mo>+</mo>
<msubsup>
<mi>f</mi>
<mi>&alpha;</mi>
<mi>v</mi>
</msubsup>
<mo>}</mo>
<mo>-</mo>
<mo>-</mo>
<mo>-</mo>
<mrow>
<mo>(</mo>
<mn>15</mn>
<mo>)</mo>
</mrow>
</mrow>
Step 5, formula (11) is utilized to obtain kth+1 strategy vk+1Performance potential vector
Step 6, judgementWhether set up, ξ is a less constant；If so, then represent obtained kth+1 time
Tactful vk+1As optimal control policy v*, and algorithm is exited, otherwise；K+1 is assigned to after k, return to step 4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610128381.0A CN105643625B (en) | 2016-03-07 | 2016-03-07 | A kind of single site based on mechanical arm is fed the method for work of production and processing system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610128381.0A CN105643625B (en) | 2016-03-07 | 2016-03-07 | A kind of single site based on mechanical arm is fed the method for work of production and processing system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105643625A CN105643625A (en) | 2016-06-08 |
| CN105643625B true CN105643625B (en) | 2017-08-25 |
Family
ID=56493203
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610128381.0A Active CN105643625B (en) | 2016-03-07 | 2016-03-07 | A kind of single site based on mechanical arm is fed the method for work of production and processing system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105643625B (en) |
Families Citing this family (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105904444B (en) * | 2016-06-24 | 2018-08-17 | 张家港市兰航机械有限公司 | A kind of operating method of mechanical arm |
| CN106965180A (en) * | 2017-04-13 | 2017-07-21 | 北京理工大学 | The mechanical arm grabbing device and method of bottle on streamline |
| CN109870992B (en) * | 2019-03-26 | 2021-09-07 | 合肥工业大学 | CSPS system control method without considering delay waiting |
| CN111517112B (en) * | 2020-04-21 | 2021-09-24 | 合肥工业大学 | Improved TP + LCM full-lamination system and control method |
| CN114193440B (en) * | 2022-01-04 | 2023-09-12 | 中船重工鹏力（南京）智能装备系统有限公司 | Robot automatic grabbing system and method based on 3D vision |
| CN114872054A (en) * | 2022-07-11 | 2022-08-09 | 深圳市麦瑞包装制品有限公司 | Method for positioning robot hand for industrial manufacturing of packaging container |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2005064205A (en) * | 2003-08-11 | 2005-03-10 | Niigata Seimitsu Kk | Transfer device and transfer method of circuit board, and method of mounting solder ball |
| US7356378B1 (en) * | 2007-04-03 | 2008-04-08 | Taiwan Semiconductor Manufacturing Company | Method and system for smart vehicle route selection |
| CN101788787B (en) * | 2009-12-31 | 2011-09-21 | 合肥工业大学 | Optimization control method of multi-site conveyor belt feeding production processing site system |
| CN102629108B (en) * | 2012-04-19 | 2013-09-18 | 合肥工业大学 | Optimization control method for multi-procedure conveyor belt feeding processing station system with flexible sites |
| CN104850117B (en) * | 2015-05-19 | 2017-05-17 | 合肥工业大学 | Robot automatic production line experiment platform and control method thereof |
-
2016
- 2016-03-07 CN CN201610128381.0A patent/CN105643625B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN105643625A (en) | 2016-06-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN105643625B (en) | A kind of single site based on mechanical arm is fed the method for work of production and processing system | |
| CN109047539B (en) | Artificial intelligence production system for sheet metal parts | |
| US10766139B2 (en) | Method for avoiding collisions between two robots | |
| WO2019056840A1 (en) | Palletizing control device, system and method and storage medium | |
| CN109789545B (en) | Method and system for controlling speed of a transport path | |
| EP3624997B1 (en) | Method and control system for controlling movement sequences of a robot | |
| CN107977738B (en) | Multi-objective optimization control method for conveyor belt feeding processing station system | |
| CN106628940A (en) | Automatic transferring system device for manufacturing air conditioning evaporator and operation method | |
| CN108197833A (en) | A kind of complete Real-time dispatch system and dispatching method towards discrete workshop | |
| CN109870992B (en) | CSPS system control method without considering delay waiting | |
| CN109500809B (en) | Optimization of an automated process for selecting and grasping objects by a robot | |
| CN111282827A (en) | Automatic workpiece feeding and quick sorting control method | |
| CN107973041A (en) | A kind of goods of Almightiness type sorts logistics device to people | |
| CN110255036B (en) | Full-automatic unattended production system | |
| CN110919655A (en) | Exoskeleton robot power auxiliary control method based on reinforcement learning | |
| CN112809678A (en) | Cooperative control method for production line system of multi-robot workstation | |
| CN205462883U (en) | Shaping sheet metal component automatic spraying unit of two ware digital bus mills in air conditioner | |
| CN110371633A (en) | A kind of automatic Material Handling System | |
| CN104859092A (en) | Intelligent production line | |
| CN104828657A (en) | Intelligent storehouse outbound device used for sheet material production line and method thereof | |
| JP2005536417A (en) | Automated production system for object identification, selection and transfer | |
| CN212190122U (en) | Automatic workpiece feeding device | |
| CN114511167A (en) | Material handling equipment scheduling method, device and system and storage medium | |
| CN111517112B (en) | Improved TP + LCM full-lamination system and control method | |
| CN114154926B (en) | Intelligent logistics system and method suitable for sheet metal part manufacturing |
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 |