CN117021118B - Dynamic compensation method for digital twin track error of parallel robot - Google Patents
Dynamic compensation method for digital twin track error of parallel robot Download PDFInfo
- Publication number
- CN117021118B CN117021118B CN202311287483.3A CN202311287483A CN117021118B CN 117021118 B CN117021118 B CN 117021118B CN 202311287483 A CN202311287483 A CN 202311287483A CN 117021118 B CN117021118 B CN 117021118B
- Authority
- CN
- China
- Prior art keywords
- speed
- digital twin
- error
- layer
- parallel robot
- 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
- 238000000034 method Methods 0.000 title claims abstract description 87
- 238000006073 displacement reaction Methods 0.000 claims abstract description 88
- 230000003993 interaction Effects 0.000 claims abstract description 43
- 230000001360 synchronised effect Effects 0.000 claims abstract description 30
- 238000012549 training Methods 0.000 claims abstract description 29
- 230000002787 reinforcement Effects 0.000 claims abstract description 15
- 238000004891 communication Methods 0.000 claims abstract description 13
- 230000008569 process Effects 0.000 claims description 32
- 230000001133 acceleration Effects 0.000 claims description 22
- 239000012636 effector Substances 0.000 claims description 17
- 238000009825 accumulation Methods 0.000 claims description 15
- 230000006870 function Effects 0.000 claims description 11
- 230000000737 periodic effect Effects 0.000 claims description 8
- 238000012545 processing Methods 0.000 claims description 7
- 230000000694 effects Effects 0.000 claims description 5
- 238000013178 mathematical model Methods 0.000 claims description 4
- 238000004458 analytical method Methods 0.000 claims description 3
- 238000005728 strengthening Methods 0.000 claims description 3
- 230000009466 transformation Effects 0.000 claims description 3
- 230000015572 biosynthetic process Effects 0.000 claims description 2
- 238000003786 synthesis reaction Methods 0.000 claims description 2
- 230000009286 beneficial effect Effects 0.000 abstract description 3
- 238000010586 diagram Methods 0.000 description 6
- 230000003068 static effect Effects 0.000 description 5
- 239000003795 chemical substances by application Substances 0.000 description 4
- 230000000452 restraining effect Effects 0.000 description 4
- 238000005516 engineering process Methods 0.000 description 3
- 230000007613 environmental effect Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000006872 improvement Effects 0.000 description 2
- 230000002452 interceptive effect Effects 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000013507 mapping Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000002829 reductive effect Effects 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000003247 decreasing effect Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000012804 iterative process Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000009751 slip forming Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
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
- B25J9/1664—Programme controls characterised by programming, planning systems for manipulators characterised by motion, path, trajectory planning
-
- 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
- B25J9/1628—Programme controls characterised by the control loop
- B25J9/163—Programme controls characterised by the control loop learning, adaptive, model based, rule based expert control
Abstract
The invention belongs to the technical field of robot running track compensation, and solves the problem of poor virtual-real interaction performance of parallel robots. The method comprises the steps of generating a digital twin model of the parallel robot and constructing a parallel robot track sample data set; training and iterating the digital twin model by a dynamic coordination reinforcement learning method to obtain an optimal motion track under optimal displacement and speed; compensating the speed error between the physical entity of the parallel robot and the optimized digital twin model by a symmetrical error compensation method; and integrating the digital twin frames to realize real-time synchronous interaction between the physical entity of the parallel robot and the digital twin model. The invention is beneficial to improving the reconstruction capability of the digital twin model and the error control performance of the parallel robot, can realize the communication synchronization between the physical entity and the digital twin model and improve the virtual-real interaction rate of the physical entity and the digital twin model.
Description
Technical Field
The invention belongs to the technical field of robot running track compensation, and particularly relates to a parallel robot digital twin track error dynamic compensation method.
Background
At present, production operations such as intelligent boxing, carrying and the like in the field of parallel robots combined with digital twin technology have become hot spots in intelligent manufacturing industry. However, for parallel robots, high-precision and high-speed operation is critical, and most digital twin system frames cannot meet the requirements of the parallel robots on high precision and high speed.
The current digital twin technology is mature in virtual-real synchronous interactive application, but the parallel robot is asynchronous in displacement during movement due to rigidity errors and assembly errors of the body; in the digital twin technology, communication delay exists between a physical entity and a digital twin model, and communication time can cause speed errors of parallel robots in synchronization, so that virtual-real interaction rate of the physical entity and the digital twin model is reduced.
Disclosure of Invention
The invention provides a dynamic compensation method for digital twin track errors of parallel robots in order to solve at least one technical problem in the prior art.
The invention is realized by adopting the following technical scheme: a dynamic compensation method for digital twin track errors of parallel robots comprises the following steps: s1: constructing a virtual model of the parallel robot, and performing rod member constraint and space constraint on the virtual model to generate a digital twin model of the parallel robot; s2: carrying out dynamic analysis on the digital twin model, collecting motor torque and kinetic energy parameters of the parallel robot, converting the motor torque and kinetic energy parameters into track displacement and speed parameters, and constructing a track sample data set of the parallel robot; s3: training and iterating the digital twin model by a dynamic coordination reinforcement learning method to obtain an optimal motion track under optimal displacement and speed; s4: compensating the speed error between the physical entity of the parallel robot and the optimized digital twin model by a symmetrical error compensation method; s5: and integrating the digital twin frames to realize real-time synchronous interaction between the physical entity of the parallel robot and the digital twin model.
Preferably, in step S2, a kinetic set of the parallel robot is constructed to obtain a trajectory displacement of the parallel robotAnd speed->The method comprises the steps of carrying out a first treatment on the surface of the And establishing a track displacement and speed mathematical model of the digital twin model, collecting parallel robot track sample data, and collecting N sample data point sets as a parallel robot track sample data set.
Preferably, the step S3 specifically includes the following steps: s31: constructing a track displacement DQN sub-network and a speed DQN sub-network: an optimal displacement model and an optimal velocity model; s32: dividing a track sample data set into track displacement and speed sample data sets, training the DQN sub-networks, introducing dynamic coordination coefficients, changing track strategies, and dynamically coordinating the displacement Q values and the speed Q values of the two DQN sub-networks through the improved optimal track strategies; s33: and constructing a DQN total network for acquiring the optimal motion trail, inputting a displacement Q value and a speed Q value, and performing training iteration on the DQN total network to finally obtain the optimal motion trail under the optimal displacement and the speed.
Preferably, in step S32, the improved optimal strategyThe formula of (2) is:
in the method, in the process of the invention,and->Dynamic coordination coefficients under the strategy; />Is an initial trajectory strategy; />Is displacement error; />Is a speed error; />Solving a function for a minimum value; />To execute the initial trajectory strategy->Is a Q value of (C).
Preferably, in step S32, the digital twin model is in a state of being physically combined with the physical entityTime displacement error->The method comprises the following steps:
in the method, in the process of the invention,representing a digital twin model at +.>Position coordinates of the end effector operation at the moment;representing physical entity in->Position coordinates of the end effector operation at the moment;
digital twin model and physical entitySpeed error +.>The method comprises the following steps:
in the method, in the process of the invention,representing that the mathematical twin model of the parallel robot is +.>Speed of end effector at time, +.>Representing that the physical entity of the parallel robot is +.>The speed of the end effector at the moment in time.
Preferably, the method comprises the steps of,as an output of the optimal displacement model and feedback acting on the optimal displacement model, the displacement error is finally output to be the smallest>As a displacement Q value; />As an output of the optimal speed model and feedback is applied to the optimal speed model, the final output is the smallest speed error +.>As the speed Q value.
Preferably, in the DQN total network, the minimum trajectory synthesis error is defined as:
in the middle ofIs the overall dynamic balance coefficient.
Preferably, there is a time delay between the physical entity of the parallel robot and the digital twin model, and in order to ensure the consistency of the operation speeds of the physical entity of the parallel robot and the digital twin model, the speed error is compensated according to a speed symmetry concept, which specifically comprises the following steps:
s41: synchronizing the operation period of the digital twin model with the time period in the physical entity movement process, wherein the operation period of the digital twin model comprises 5 links which are sequentially: starting an acceleration link, a uniform acceleration to uniform deceleration link, a uniform deceleration link and a final braking deceleration link; dividing the same operation period by the physical entities of the parallel robot; s42: enabling the physical entity to perform periodic motion according to the optimal motion trail of the digital twin model, and setting a state node to correspond to the gate-type trail motion; s43: the gate-type motion trail is in a symmetrical shape, and motion with consistent transformation rule in the period is divided by adopting a symmetrical error compensation method; the uniform deceleration link and the final braking deceleration link are simplified into speed symmetrical models of the uniform acceleration link and the starting acceleration link, the center line is taken as a reference, the left side is an acceleration trend speed model, the right side is a deceleration trend speed model, and meanwhile, the speed strategy of the digital twin model and the physical entity is changed, so that the movement trend of the digital twin model and the physical entity is consistent; s44: in the starting acceleration link, the uniform acceleration to uniform deceleration link and the final braking deceleration link, the error compensation is linear error compensation; in the uniform acceleration link and the uniform deceleration link, the error compensation is arc error compensation.
Preferably, in step S5, the real-time synchronous interaction system of the parallel robot physical entity and the digital twin model includes: the system comprises a physical entity layer, an information interaction layer, an error feedback layer, a data accumulation layer, a virtual twin layer, a data processing layer, a dynamic coordination layer, a virtual-real interaction module and a dynamic coordination reinforcement learning module;
the physical entity layer comprises a parallel robot body and a communication port; the information interaction layer is used for controlling communication linkage between the running software and the digital twin model; the error feedback layer is used for eliminating the synchronous error of the optimal track displacement and the optimal track speed of the physical entity layer and the virtual twin layer; data accumulation establishes database information, and saves training parameters of the DQN network and compensation parameters in the parallel robot motion process; the virtual twin layer comprises a parallel robot digital twin model, a software platform interface and a mobile control system; the dynamic coordination layer coordinates the current track error and the speed error in combination with the optimal strategy in dynamic coordination reinforcement learning, and is responsible for feedback coordination of the physical entity layer and the virtual twin layer to form an overall closed-loop interaction system.
Preferably, the real-time synchronous interaction process of the parallel robot physical entity and the digital twin model is as follows:
starting a physical entity layer to enable the parallel robot to do periodic motion, and entering an information interaction layer through a communication interface; starting a virtual twin layer, dynamically coordinating and strengthening the learning module to perform optimal track displacement, and removing redundant data information in a data processing layer;
the information interaction layer analyzes the kinematic parameters in the physical entity layer, converts the kinematic parameters into digital information and transmits the digital information to the error feedback layer; meanwhile, the dynamic coordination reinforcement learning module calculates and accumulates the rewarding value and transmits the current data to the data accumulation layer and the error feedback layer;
the error feedback layer analyzes the synchronous result, achieves synchronous compensation of the track speed on the basis of synchronous track displacement, obtains smaller speed and displacement error, and achieves synchronous control effect of the high-precision digital twin model and the object entity;
and finally, the historical information in the data accumulation layer and the information in the current error feedback layer are dynamically coordinated, and are fed back to the physical entity layer and the virtual twin layer in a closed loop mode to complete the virtual-real interaction process.
Compared with the prior art, the invention has the beneficial effects that:
the invention constructs a virtual model of the parallel robot and generates a digital twin model of the parallel robot; collecting torque and kinetic energy parameters of the parallel robot and converting the torque and kinetic energy parameters into track displacement and speed parameters; training and iterating the digital twin model by a dynamic coordination reinforcement learning method to obtain the minimum displacement error and the minimum speed error under the optimal motion track; compensating the speed error between the physical entity of the parallel robot and the optimized digital twin model by a symmetrical error compensation method; and integrating the digital twin frames to realize real-time synchronous interaction between the physical entity of the parallel robot and the digital twin model. The method is beneficial to improving the reconstruction capability of the digital twin model and the error control performance of the parallel robot, can realize communication synchronization between a physical entity and the digital twin model, and improves the virtual-real interaction rate of the physical entity and the digital twin model.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are needed in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and other drawings can be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic overall flow diagram of the present invention;
FIG. 2 is a flow chart of an implementation of the CD-DQN network of the present invention;
FIG. 3 is a flow chart of an implementation of a DQN network prior to modification;
FIG. 4 is a graph comparing the velocity error of a CD-DQN network of the present invention with a DQN network prior to improvement;
FIG. 5 is a schematic diagram of a trajectory displacement versus velocity mathematical model of the present invention;
FIG. 6 is a schematic diagram of a symmetrical error compensation structure of the present invention;
fig. 7 is a block diagram of the dynamic virtual-real interaction method of the present invention;
FIG. 8 is a block diagram of a prior art virtual-to-real interaction method;
fig. 9 is a diagram showing comparison of interaction synchronization rates of the dynamic virtual-real interaction method and the virtual-real interaction method before improvement.
Detailed Description
Technical solutions in the embodiments of the present invention will be clearly and completely described with reference to the drawings in the embodiments of the present invention, and it is apparent that the described embodiments are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the examples of this invention without making any inventive effort, are intended to fall within the scope of this invention.
It should be understood that the structures, proportions, sizes, etc. shown in the drawings are merely for the purpose of understanding and reading the disclosure, and are not intended to limit the scope of the invention, which is defined by the appended claims, and any structural modifications, proportional changes, or dimensional adjustments, which may be made by those skilled in the art, should fall within the scope of the present disclosure without affecting the efficacy or the achievement of the present invention, and it should be noted that, in the present disclosure, relational terms such as first and second are used solely to distinguish one entity from another entity without necessarily requiring or implying any actual relationship or order between such entities.
The present invention provides an embodiment:
as shown in fig. 1, a dynamic compensation method for digital twin track errors of a parallel robot includes the following steps:
s1: constructing a virtual model of the parallel robot, and performing rod member constraint and space constraint on the virtual model to generate a digital twin model of the parallel robot;
s2: carrying out dynamic analysis on the digital twin model, collecting motor torque and kinetic energy parameters of the parallel robot, converting the motor torque and kinetic energy parameters into track displacement and speed parameters, and collecting a track sample data set of the parallel robot;
s3: training and iterating the digital twin model by a dynamic coordination reinforcement learning method to obtain an optimal motion track under optimal displacement and speed;
s4: compensating the speed error between the physical entity of the parallel robot and the optimized digital twin model by a symmetrical error compensation method;
s5: and integrating the digital twin frames to realize real-time synchronous interaction between the physical entity of the parallel robot and the digital twin model.
In step S1, a three-dimensional model of the parallel robot is built, modular division is performed, the three-dimensional model comprises a static platform, a dynamic platform, a driving rod, a spring and a screw part, and main components of the three-dimensional model are assembled, so that a virtual model of the parallel robot is built. And fixing and restraining the virtual model of the parallel robot by using a physical engine, setting the working space of the parallel robot, setting the space working range of the parallel robot, and establishing a digital twin model of the parallel robot.
The method comprises the following specific steps: fixing and restraining the parallel robot by using a Unity3D physical engine, and registering a static platform coordinate system in the parallel robot with an actual world coordinate system; fixing a static platform part of the parallel robot, performing spherical hinge constraint on the driving rod piece, and performing parallel constraint on a dynamic platform part of the parallel robot; restraining the structural relation between the driving rod piece and the movable platform to finish fixing and restraining operation; the method comprises the steps of endowing the movable platform part with stress with the same size in different directions, setting a normal working space of a parallel robot in a space coordinate system where a static platform is positioned, and setting a reasonable working range; and adding the structural mapping relation of the parallel robot to generate a digital twin model of the parallel robot.
In step S2, the kinematic set constructed by the physical information of the parallel robot is as followsCollecting all parameters of the parallel robot, converting the parameters into kinematic information of the parallel robot, and passing through the radius of the static platform>Radius of movable platform->Active lever length->Driven rod length->Obtaining the position and posture relation of the tail ends of the parallel robots; the track displacement of the parallel robot is obtained through the constructed kinematic model>And speed->The method comprises the steps of carrying out a first treatment on the surface of the And (3) using an ML-agent plug-in the Unity3D to enable a digital twin model of the parallel robot to perform motion in a virtual space, collecting parallel robot track sample data, and collecting 50000 sample data point sets as a parallel robot track sample data set.
As shown in fig. 5, the digital twin model is run-time to include 5 links, which are defined as a trajectory displacement and velocity mathematical model:
corresponding run period；
、/>And->Respectively express +.>Acceleration, velocity and displacement at time; />Is the maximum acceleration during the run period.
Definition of the definitionThe interval corresponds to the start acceleration segment, corresponds to segment 1 in fig. 5, and is within this interval:
、/>、
；
definition of the definitionThe interval corresponds to the ramp up link, corresponding to link 2 in fig. 5, within this interval:
、/>、
；
definition of the definitionThe interval corresponds to the ramp up to ramp down link, corresponding to link 3 in fig. 5, within this interval:
、/>、
definition of the definitionThe interval corresponds to a uniform deceleration link, pairShould link 4 in fig. 5, within this interval:
、/>、
definition of the definitionThe interval corresponds to the final braking deceleration segment, corresponding to segment 5 in fig. 5, within this interval:
、/>、
；
in step S3, the digital twin model is iteratively trained by using sample data sets of parallel robot tracks, two DQN sub-networks are constructed for solving the optimal displacement and the optimal speed, and the sample data sets of the tracks are divided into sample data sets of the displacement and the speed through a dynamic coordination process. Model training of the two DQN subnetworks is respectively carried out, track displacement and speed are dynamically changed in the training process, displacement and speed errors in a digital twin model under the optimal track are solved, and the displacement Q value and the speed Q value are obtained when the errors are minimum.
And then, continuously testing errors in the DQN total network, wherein the error states comprise operation disorder and singular pose of the robot end effector, and finally, obtaining the optimal gate-type operation track while obtaining the optimal displacement and speed. The CD-DQN network is composed of two DQN sub-networks and one DQN total network, and is characterized in that the two sub-networks are biased towards the possibility of exploring track operation, while the total network reduces the exploring and simultaneously biased towards the execution activity of the optimal track, the exploring and the execution work are performed in a serial mode, and the coordination relation among multiple parameters (speed and displacement) is focused.
As shown in fig. 2 to 4, the step S3 specifically includes the following steps:
s31: two DQN subnetworks are built: an optimal displacement model A1 and an optimal velocity model B1; and defining various parameters of the coordinated dynamic network, wherein the agent is an intelligent executive and is used for executing a converged target strategy.
Specifically, S311: two different learning environments are defined respectively: definition of the definitionIs->Time-of-day corresponding global environmental status,/->、/>Respectively->The environment states of the optimal displacement model and the optimal speed model corresponding to the moment;
s312: definition of respectively、/>Is->Actions executable in the moment optimal displacement model and the optimal speed model;
s313: defining a reward mechanism，/>Representing a total prize value; at->Total environmental state at timeAnd action->、/>The total prize value obtained is +.>，/>For the magnitude of the displacement reward value +.>The magnitude of the prize value for the speed;
s314: digital twin model inThe current position of the end effector operation at the moment is +.>The physical entity of the parallel robot is +.>The current position of the end effector operation at the moment is +.>The method comprises the steps of carrying out a first treatment on the surface of the Digital twin model and physical entity are in->The displacement error at the moment is:
in the method, in the process of the invention,representing a digital twin model at +.>Position coordinates of the end effector operation at the moment;representing physical entity in->Position coordinates of the end effector operation at the moment; />Representing displacement errors.
Parallel robot mathematical twin modelThe end effector speed at time is +.>The physical entity of the parallel robot is +.>Speed of end effector at time +.>The digital twin model is +.>The speed error at the moment is:
in the method, in the process of the invention,indicating a speed error.
S315: the initial trajectory strategy function isIn the formula->To execute the initial trajectory strategy->Q value of>As a function of the minimum value.
S32: introducing a dynamic coordination coefficient and changing a track strategy; dynamically coordinating the target Q values of the two DQN subnetworks through the improved optimal strategy under the optimal track state, and dynamically coordinating the relationship between the two DQN subnetworks; the initial track is the track strategy of S315, and moves according to the correct gate track in the process of moving the track, and the track strategy function changes to become the next trackThe change in the speed and displacement error starts to be emphasized, and the speed and error becomes optimal when it is minimum.
Optimal strategy in dynamic coordination processThe formula is:
in the method, in the process of the invention,and->Dynamic coordination coefficients under the strategy; />Is an initial trajectory strategy; />Is displacement error; />Is a speed error; />To execute the initial trajectory strategy->Is a Q value of (C).
When the speed error of the digital twin model is larger than 0.03m/s and the displacement error is smaller than 0.2mm, attention is paid to the minimization of the displacement error; when the speed error of the digital twin model is smaller than 0.03m/s and the displacement error is larger than 0.2mm, paying attention to the speed error minimization; the method has the advantages that the kinematic model with larger error is eliminated, and a better network evaluation effect can be obtained.
If both conform to the basic error, the initial trajectory strategy is kept unchanged; if the running speed and the track error exceed fixed values, the strategy value is set to 0, and the accumulation and calculation of the rewarding value are not carried out, so that the network calculated amount is reduced.
Training subnetworks: training of the CD-DQN network is performed,as an output of the optimal displacement model and feedback acting on the optimal displacement model, the displacement error is finally output to be the smallest>As a displacement Q value; />As an output of the optimal speed model and feedback is applied to the optimal speed model, the final output is the smallest speed error +.>As a speed Q value; the original 50000 sample points are integrated into 15000 more optimal track samples, and the displacement Q value and the speed Q value are calculatedAs an input Q value of the overall DQN network.
S33: constructing a DQN total network for acquiring an optimal motion trail, and inputting a displacement Q value and a speed Q value to carry out training iteration; the DQN total network training process obtains a running track with the minimum error;
specifically, S331: firstly, performing DQN total network training on a data sample set and off-line strategyCompleting exploration and learning, and sampling a plurality of batches of samples; defining the minimum track integrated error as:
in the formula->Is the overall dynamic balance coefficient;
the total prize value isIn the formula->For the minimum of the integrated errors of the trajectories,as a function of the maximum value.
S332: the optimal track obtained by the displacement error and the speed error needs to be subjected to periodic motion when the parallel robot runs; running period of digital twin model of parallel robotFurther divide into->A status node with a time interval of 20ms, here +.>，/>Each state node is a further division in 5 links. The sample number after the sub-network simplification is 15000, and 5000 times of iterative computation are started;
s333: solving a loss function of the DQN total network;
in the method, in the process of the invention,mathematical expectations for empirical playback; />For total weight, ++>Weights in the displacement and velocity subnetworks, respectively; />，/>，/>To approximate the input Q function, < >>Is->Time of day global environmental status->Strategy +.>Is a function of the Q value of (c).
S334: training the overall DQN network, and firstly, in the weight alternation process, the target Q value is calculatedFixing is performed, and then every 10 iterative training sessions will be followed by an assessment of the updated total weight in the network +.>Endow->Preliminary iterations were performed, with 5000 times of overall network training. The calculation formula of the target Q value is as follows:
in the method, in the process of the invention,to perform->Current optimal Q value of policy, +.>For a current optimal Q function approaching an optimal strategy, < ->Is a mathematical expectation.
Specifically, optimal displacement and optimal speed parameters are obtained in two DQN subnetworks, and the optimal displacement and the optimal speed parameters are substituted into a total network as input Q values to obtain an optimal track route of the robot. Dividing the displacement sample parameters and the speed sample parameters in the track sample data set obtained in the step S2, and changing the number of samples from 50000 to 15000 through two DQN sub-networks of an optimal displacement model and an optimal speed model. Setting the optimal running track as the DQN total networkThe obtained speed Q value and displacement Q value of the DQN sub-network are taken as the input Q value of the total network; under the optimal strategy, the DQN total network can predict the optimal Q value in real time, and the optimal Q value is compared with the optimal Q value/>Comparison is performed such that the loss function in the iterative process +.>Converging; in the training iteration process, the track sample set data is simplified, the end effector of the robot digital twin model predicts various tracks in the motion process, and the track sample is simplified; input Q value which is updated every 10 times training is performed in network iteration, and the input Q value is weighted +.>Giving a target Q value, and finishing updating the target Q value; new input Q values and target Q values are continuously formed in the network training iteration to achieve dynamic track error compensation.
Converting the original kinematic parameters into track coordinate parameters through a digital twin model; the ML-agent plug-in the Unity3D engine is used for adopting an experience playback principle, namely the agent continuously performs interactive data iteration to form a parameter setFor->Total prize value +.>Accumulation is carried out, and in the training process, the environment is->Make a change to->Obtaining the minimum comprehensive error, and finally obtaining the synchronization of the optimal path and the end effector of the robot digital twin model by synchronous mapping; after the speed and track displacement compensation adjustment, the running in the digital twin model is synchronous with the optimal track, and the minimum comprehensive error is obtained. The smaller the error is, the higher the prize value is, and the synchronization is completedThen add->、/>The next training iteration is performed in the return experience playback.
After training, the digital twin model data are updated in real time, and virtual-real synchronization of the optimal motion trail and the motion trail of the twin model is obtained. The quality of the network training is detected, and the speed error of the original DQN is compared with that of the improved CD-DQN. As shown in FIG. 5, 60 seconds was chosen as the motion period, the trajectory was repeatedly cycled, and the CD-DQN method was significantly better than the original DQN method in that the velocity error decreased faster over time.
As shown in fig. 5 and fig. 6, after the dynamic reinforcement learning process is completed, the digital twin model of the parallel robot completes the operation of the optimal track displacement and speed, and performs periodic motion in the virtual space, and as the physical entity and the digital twin model have time delay, the actual error compensation of the physical entity and the digital twin model of the parallel robot is required, and a speed symmetry concept is provided, so that the motion trend of the digital twin model is consistent with the operation speed of the physical entity.
Specifically, S41: synchronizing the operation period of the digital twin model with the time period in the physical entity movement process, wherein the operation period of the digital twin model is as followsDividing the physical entities of the parallel robots into the same operation period;
s42: enabling the physical entity to perform periodic motion according to the optimal displacement track of the digital twin model, and setting a state node to correspond to the gate-type track motion;
s43: the gate-type motion trail is in a symmetrical shape, and motion with consistent transformation rule in the period is divided by adopting a symmetrical error compensation method; simplifying the uniform deceleration link and the final braking deceleration link into speed symmetric models of the uniform acceleration link and the starting acceleration link (speed model simplification is carried out on the link 4 and the link 5, the link 4 is changed into the link 2', the link 5 is changed into the 1'. 1 'and the link 2' are speed symmetric models of the links 1 and 2), taking the central line as a reference, taking the left side as an acceleration trend speed model, taking the right side as a deceleration trend speed model, and simultaneously changing the speed strategy in the digital twin model and the physical entity, so that the movement trend of the digital twin model and the physical entity is consistent;
s44: the specific compensation method comprises the following steps:、/>、/>when the error compensation is mainly linear error compensation; />、/>In this case, the error compensation is mainly arc error compensation.
As shown in fig. 7 to 9, according to the optimal track executed by the parallel robot trained in the step S3 and the feedback error compensation obtained in the step S4, digital twin frame integration is performed, and real-time synchronous interaction between the physical entity of the parallel robot and the digital twin model is realized by setting a physical entity layer, a virtual twin layer, an information interaction layer, an error feedback layer, a data processing layer, a data accumulation layer and a dynamic coordination layer.
The real-time synchronous interaction system comprises: the system comprises a physical entity layer, an information interaction layer, an error feedback layer, a data accumulation layer, a virtual twin layer, a data processing layer, a dynamic coordination layer, a virtual-real interaction module and a dynamic coordination reinforcement learning module; the physical entity layer comprises a parallel robot body and a communication port; the information interaction layer is used for controlling communication linkage between the running software and the digital twin model; the error feedback layer is used for eliminating the synchronous error of the optimal track displacement and the optimal track speed of the physical entity layer and the virtual twin layer; the data accumulation layer establishes database information and stores training parameters of the DQN network and compensation parameters in the parallel robot motion process; the virtual twin layer comprises a parallel robot digital twin model, a software platform interface and a mobile control system; the dynamic coordination layer coordinates the current track error and the speed error in combination with the optimal strategy in dynamic coordination reinforcement learning, and is responsible for feedback coordination of the physical entity layer and the virtual twin layer to form an overall closed-loop interaction system.
The real-time synchronous interaction process of the physical entity of the parallel robot and the digital twin model comprises the following steps:
starting a physical entity layer to enable the robot to do periodic motion, and entering an information interaction layer through a communication interface; starting a virtual twin layer, dynamically coordinating and strengthening the learning module to perform optimal track displacement, and removing redundant data information in a data processing layer;
the information interaction layer analyzes dynamic parameters in the physical entity layer, converts the dynamic parameters into digital information and transmits the digital information to the error feedback layer; meanwhile, the dynamic coordination reinforcement learning module calculates and accumulates the rewarding value and transmits the current data to the data accumulation layer and the error feedback layer;
the error feedback layer analyzes the synchronous result, achieves synchronous compensation of the track speed on the basis of synchronous track displacement, obtains smaller speed and displacement error, and achieves synchronous control effect of the high-precision digital twin model and the object entity;
and finally, the historical information in the data accumulation layer and the information in the current error feedback layer are dynamically coordinated, and are fed back to the physical entity layer and the virtual twin layer in a closed loop mode to complete the virtual-real interaction process.
The foregoing is only a preferred embodiment of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions easily contemplated by those skilled in the art within the technical scope of the present invention should be included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (2)
1. The dynamic compensation method for the digital twin track error of the parallel robot is characterized by comprising the following steps of:
s1: constructing a virtual model of the parallel robot, and performing rod member constraint and space constraint on the virtual model to generate a digital twin model of the parallel robot;
s2: carrying out dynamic analysis on the digital twin model, collecting motor torque and kinetic energy parameters of the parallel robot, converting the motor torque and kinetic energy parameters into track displacement and speed parameters, and constructing a track sample data set of the parallel robot;
s3: training and iterating the digital twin model by a dynamic coordination reinforcement learning method to obtain an optimal motion track under optimal displacement and speed;
s4: compensating the speed error between the physical entity of the parallel robot and the optimized digital twin model by a symmetrical error compensation method;
s5: integrating the digital twin frames to realize real-time synchronous interaction between the physical entity of the parallel robot and the digital twin model;
step S3 comprises the steps of:
s31: constructing a track displacement DQN sub-network and a speed DQN sub-network: an optimal displacement model and an optimal velocity model;
s32: dividing a track sample data set into track displacement and speed sample data sets, training the DQN sub-networks, introducing dynamic coordination coefficients, changing track strategies, and dynamically coordinating the displacement Q values and the speed Q values of the two DQN sub-networks through the improved optimal track strategies;
s33: constructing a DQN total network for acquiring an optimal motion trail, inputting a displacement Q value and a speed Q value, and performing training iteration on the DQN total network to finally acquire the optimal motion trail under the optimal displacement and the speed;
in step S32, the improved optimal strategyThe formula of (2) is:
in the method, in the process of the invention,and->Dynamic coordination coefficients under the strategy; />Is an initial trajectory strategy; />Is displacement error; />Is a speed error; />Solving a function for a minimum value; />To execute the initial trajectory strategy->Q value of (2);
in step S32, the digital twin model and the physical entity are inTime displacement error->The method comprises the following steps:
in the method, in the process of the invention,representing a digital twin model at +.>Position coordinates of the end effector operation at the moment; />Representing physical entity in->Position coordinates of the end effector operation at the moment;
digital twin model and physical entitySpeed error +.>The method comprises the following steps:
in the method, in the process of the invention,representing that the mathematical twin model of the parallel robot is +.>Speed of end effector at time, +.>Representing that the physical entity of the parallel robot is +.>The speed of the end effector at the moment;
as an output of the optimal displacement model and feedback acting on the optimal displacement model, the displacement error is finally output to be the smallest>As a displacement Q value; />As an output of the optimal speed model and feedback is applied to the optimal speed model, the final output is the smallest speed error +.>As a speed Q value;
in step S33, in the DQN total network, the minimum trajectory synthesis error is defined as:
in the middle ofIs the overall dynamic balance coefficient;
in order to ensure the consistency of the two operation speeds, the speed error is compensated according to the speed symmetry concept, and the step S4 comprises the following steps:
s41: synchronizing the operation period of the digital twin model with the time period in the physical entity movement process, wherein the operation period of the digital twin model comprises 5 links which are sequentially: starting an acceleration link, a uniform acceleration to uniform deceleration link, a uniform deceleration link and a final braking deceleration link; dividing the same operation period by the physical entities of the parallel robot;
s42: enabling the physical entity to perform periodic motion according to the optimal motion trail of the digital twin model, and setting a state node to correspond to the gate-type trail motion;
s43: the gate-type motion trail is in a symmetrical shape, and motion with consistent transformation rule in the period is divided by adopting a symmetrical error compensation method; the uniform deceleration link and the final braking deceleration link are simplified into speed symmetrical models of the uniform acceleration link and the starting acceleration link, the center line is taken as a reference, the left side is an acceleration trend speed model, the right side is a deceleration trend speed model, and meanwhile, the speed strategy of the digital twin model and the physical entity is changed, so that the movement trend of the digital twin model and the physical entity is consistent;
s44: in the starting acceleration link, the uniform acceleration to uniform deceleration link and the final braking deceleration link, the error compensation is linear error compensation; in the uniform acceleration link and the uniform deceleration link, the error compensation is arc error compensation;
in step S5, the real-time synchronous interaction system of the parallel robot physical entity and the digital twin model includes: the system comprises a physical entity layer, an information interaction layer, an error feedback layer, a data accumulation layer, a virtual twin layer, a data processing layer, a dynamic coordination layer, a virtual-real interaction module and a dynamic coordination reinforcement learning module; the physical entity layer comprises a parallel robot body and a communication port; the information interaction layer is used for controlling communication linkage between the running software and the digital twin model; the error feedback layer is used for eliminating the synchronous error of the optimal track displacement and the optimal track speed of the physical entity layer and the virtual twin layer; data accumulation establishes database information, and saves training parameters of the DQN network and compensation parameters in the parallel robot motion process; the virtual twin layer comprises a parallel robot digital twin model, a software platform interface and a mobile control system; the dynamic coordination layer coordinates the current track error and the speed error in combination with the optimal strategy in dynamic coordination reinforcement learning, and is responsible for the feedback coordination of the physical entity layer and the virtual twin layer to form an overall closed-loop interaction system;
the real-time synchronous interaction process of the physical entity of the parallel robot and the digital twin model comprises the following steps: starting a physical entity layer to enable the parallel robot to do periodic motion, and entering an information interaction layer through a communication interface; starting a virtual twin layer, dynamically coordinating and strengthening the learning module to perform optimal track displacement, and removing redundant data information in a data processing layer; the information interaction layer analyzes the kinematic parameters in the physical entity layer, converts the kinematic parameters into digital information and transmits the digital information to the error feedback layer; meanwhile, the dynamic coordination reinforcement learning module calculates and accumulates the rewarding value and transmits the current data to the data accumulation layer and the error feedback layer; the error feedback layer analyzes the synchronous result, achieves synchronous compensation of the track speed on the basis of synchronous track displacement, obtains smaller speed and displacement error, and achieves synchronous control effect of the high-precision digital twin model and the object entity; and finally, the historical information in the data accumulation layer and the information in the current error feedback layer are dynamically coordinated, and are fed back to the physical entity layer and the virtual twin layer in a closed loop mode to complete the virtual-real interaction process.
2. The method for dynamically compensating digital twin track errors of parallel robot according to claim 1, wherein the method comprises the following steps: in step S2, a parallel robot dynamics set is constructed to obtain the track displacement of the parallel robotAnd speed->The method comprises the steps of carrying out a first treatment on the surface of the And establishing a track displacement and speed mathematical model of the digital twin model, collecting parallel robot track sample data, and collecting N sample data point sets as a parallel robot track sample data set.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311287483.3A CN117021118B (en) | 2023-10-08 | 2023-10-08 | Dynamic compensation method for digital twin track error of parallel robot |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311287483.3A CN117021118B (en) | 2023-10-08 | 2023-10-08 | Dynamic compensation method for digital twin track error of parallel robot |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117021118A CN117021118A (en) | 2023-11-10 |
| CN117021118B true CN117021118B (en) | 2023-12-15 |
Family
ID=88634052
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311287483.3A Active CN117021118B (en) | 2023-10-08 | 2023-10-08 | Dynamic compensation method for digital twin track error of parallel robot |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117021118B (en) |
Citations (24)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH02152777A (en) * | 1988-12-02 | 1990-06-12 | Meidensha Corp | Control method for bilateral master/slave manipulator |
| JPH05237773A (en) * | 1992-02-26 | 1993-09-17 | Tokyo Electric Power Co Inc:The | Master/slave type manipulator |
| CN103092203A (en) * | 2013-01-15 | 2013-05-08 | 深圳市紫光杰思谷科技有限公司 | Control method of relative motion between primary robot and secondary robot |
| KR101448465B1 (en) * | 2013-05-20 | 2014-10-13 | 한국생산기술연구원 | Unified Motion-Based Bilateral Teleoperation Controller and Bilateral Teleoperation Control method |
| JP2015037824A (en) * | 2013-08-19 | 2015-02-26 | 学校法人立命館 | Master-slave system |
| DE102015116086A1 (en) * | 2015-09-23 | 2017-03-23 | Universität Bayreuth | robot control |
| JP2019155522A (en) * | 2018-03-12 | 2019-09-19 | オムロン株式会社 | Control device, control method, and program |
| CN111596614A (en) * | 2020-06-02 | 2020-08-28 | 中国科学院自动化研究所 | Motion control error compensation system and method based on cloud edge cooperation |
| CN112698625A (en) * | 2020-12-11 | 2021-04-23 | 西安交通大学 | Five-axis machine tool digital twin modeling method for linkage trajectory error prediction |
| CN112859739A (en) * | 2021-01-15 | 2021-05-28 | 天津商业大学 | Digital twin-driven multi-axis numerical control machine tool contour error suppression method |
| CN113021330A (en) * | 2019-12-24 | 2021-06-25 | 沈阳智能机器人创新中心有限公司 | Multi-robot synchronous follow-up control method under distributed network |
| CN113547196A (en) * | 2021-07-27 | 2021-10-26 | 天津大学 | Master-slave mirror image synchronous motion control method of bilateral friction stir welding equipment |
| CN113687659A (en) * | 2021-10-26 | 2021-11-23 | 武汉鼎元同立科技有限公司 | Optimal trajectory generation method and system based on digital twinning |
| CN113742964A (en) * | 2021-08-19 | 2021-12-03 | 三峡大学 | Digital twinning construction method applied to numerical control machine tool |
| CN113878581A (en) * | 2021-10-28 | 2022-01-04 | 天津大学 | Error prediction and real-time compensation technology for five-degree-of-freedom hybrid robot |
| WO2022007753A1 (en) * | 2020-07-06 | 2022-01-13 | 北京卫星制造厂有限公司 | Digital twin modeling method oriented to mobile robot milling processing |
| CN114564009A (en) * | 2022-01-21 | 2022-05-31 | 首都医科大学 | Surgical robot path planning method and system |
| FR3117393A1 (en) * | 2020-12-16 | 2022-06-17 | Wandercraft | Process for setting an exoskeleton in motion |
| CN115338869A (en) * | 2022-09-13 | 2022-11-15 | 浙江大学 | Master-slave control method and system for master-slave heterogeneous teleoperation system |
| CN115556112A (en) * | 2022-10-28 | 2023-01-03 | 北京理工大学 | Robot teleoperation method and system based on digital twins |
| US11577834B1 (en) * | 2021-11-05 | 2023-02-14 | Guangdong University Of Technology | Method and system for synchronizing virtual and real statuses of digital twin system of unmanned aerial vehicle (UAV) |
| CN116300691A (en) * | 2023-05-25 | 2023-06-23 | 深圳市正和楚基科技有限公司 | State monitoring method and system for multi-axis linkage numerical control machining |
| KR20230123060A (en) * | 2022-02-14 | 2023-08-23 | 주식회사 넥스트디앤에이 | Robot monitoring apparatus and robot monitoring method |
| CN116834009A (en) * | 2023-07-19 | 2023-10-03 | 华中科技大学 | Online compensation system for pose errors of distributed virtual-real twin robot |
Family Cites Families (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8423186B2 (en) * | 2009-06-30 | 2013-04-16 | Intuitive Surgical Operations, Inc. | Ratcheting for master alignment of a teleoperated minimally-invasive surgical instrument |
| KR100968944B1 (en) * | 2009-12-14 | 2010-07-14 | (주) 아이알로봇 | Apparatus and method for synchronizing robot |
| JP6201126B2 (en) * | 2013-11-07 | 2017-09-27 | 株式会社人機一体 | Master-slave system |
| WO2020055903A1 (en) * | 2018-09-10 | 2020-03-19 | Fanuc America Corporation | Robot calibration for ar and digital twin |
| US11312012B2 (en) * | 2019-01-01 | 2022-04-26 | Giant Ai, Inc. | Software compensated robotics |
| EP3911478A1 (en) * | 2019-01-16 | 2021-11-24 | Telefonaktiebolaget Lm Ericsson (Publ) | System for emulating remote control of a physical robot |
| US20220402131A1 (en) * | 2021-06-09 | 2022-12-22 | eBots Inc. | System and method for error correction and compensation for 3d eye-to-hand coordinaton |
| US20220413500A1 (en) * | 2021-06-29 | 2022-12-29 | Ben Eazzetta | System and Method for Robotic Mission Planning & Routing |
-
2023
- 2023-10-08 CN CN202311287483.3A patent/CN117021118B/en active Active
Patent Citations (24)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JPH02152777A (en) * | 1988-12-02 | 1990-06-12 | Meidensha Corp | Control method for bilateral master/slave manipulator |
| JPH05237773A (en) * | 1992-02-26 | 1993-09-17 | Tokyo Electric Power Co Inc:The | Master/slave type manipulator |
| CN103092203A (en) * | 2013-01-15 | 2013-05-08 | 深圳市紫光杰思谷科技有限公司 | Control method of relative motion between primary robot and secondary robot |
| KR101448465B1 (en) * | 2013-05-20 | 2014-10-13 | 한국생산기술연구원 | Unified Motion-Based Bilateral Teleoperation Controller and Bilateral Teleoperation Control method |
| JP2015037824A (en) * | 2013-08-19 | 2015-02-26 | 学校法人立命館 | Master-slave system |
| DE102015116086A1 (en) * | 2015-09-23 | 2017-03-23 | Universität Bayreuth | robot control |
| JP2019155522A (en) * | 2018-03-12 | 2019-09-19 | オムロン株式会社 | Control device, control method, and program |
| CN113021330A (en) * | 2019-12-24 | 2021-06-25 | 沈阳智能机器人创新中心有限公司 | Multi-robot synchronous follow-up control method under distributed network |
| CN111596614A (en) * | 2020-06-02 | 2020-08-28 | 中国科学院自动化研究所 | Motion control error compensation system and method based on cloud edge cooperation |
| WO2022007753A1 (en) * | 2020-07-06 | 2022-01-13 | 北京卫星制造厂有限公司 | Digital twin modeling method oriented to mobile robot milling processing |
| CN112698625A (en) * | 2020-12-11 | 2021-04-23 | 西安交通大学 | Five-axis machine tool digital twin modeling method for linkage trajectory error prediction |
| FR3117393A1 (en) * | 2020-12-16 | 2022-06-17 | Wandercraft | Process for setting an exoskeleton in motion |
| CN112859739A (en) * | 2021-01-15 | 2021-05-28 | 天津商业大学 | Digital twin-driven multi-axis numerical control machine tool contour error suppression method |
| CN113547196A (en) * | 2021-07-27 | 2021-10-26 | 天津大学 | Master-slave mirror image synchronous motion control method of bilateral friction stir welding equipment |
| CN113742964A (en) * | 2021-08-19 | 2021-12-03 | 三峡大学 | Digital twinning construction method applied to numerical control machine tool |
| CN113687659A (en) * | 2021-10-26 | 2021-11-23 | 武汉鼎元同立科技有限公司 | Optimal trajectory generation method and system based on digital twinning |
| CN113878581A (en) * | 2021-10-28 | 2022-01-04 | 天津大学 | Error prediction and real-time compensation technology for five-degree-of-freedom hybrid robot |
| US11577834B1 (en) * | 2021-11-05 | 2023-02-14 | Guangdong University Of Technology | Method and system for synchronizing virtual and real statuses of digital twin system of unmanned aerial vehicle (UAV) |
| CN114564009A (en) * | 2022-01-21 | 2022-05-31 | 首都医科大学 | Surgical robot path planning method and system |
| KR20230123060A (en) * | 2022-02-14 | 2023-08-23 | 주식회사 넥스트디앤에이 | Robot monitoring apparatus and robot monitoring method |
| CN115338869A (en) * | 2022-09-13 | 2022-11-15 | 浙江大学 | Master-slave control method and system for master-slave heterogeneous teleoperation system |
| CN115556112A (en) * | 2022-10-28 | 2023-01-03 | 北京理工大学 | Robot teleoperation method and system based on digital twins |
| CN116300691A (en) * | 2023-05-25 | 2023-06-23 | 深圳市正和楚基科技有限公司 | State monitoring method and system for multi-axis linkage numerical control machining |
| CN116834009A (en) * | 2023-07-19 | 2023-10-03 | 华中科技大学 | Online compensation system for pose errors of distributed virtual-real twin robot |
Non-Patent Citations (3)
| Title |
|---|
| 基于NSGA-II算法的并联机器人结构分析及优化;范浩东等;《机械设计与制造工程》;全文 * |
| 基于RBF神经网络的多移动机械臂同步控制;张佳舒;赵阁;高守礼;赵东亚;;山东科技大学学报(自然科学版)(第03期);全文 * |
| 基于马氏距离的孪生多分类支持向量机;张谢锴;丁世飞;;计算机科学(第03期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117021118A (en) | 2023-11-10 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Rodriguez et al. | DeepWalk: Omnidirectional bipedal gait by deep reinforcement learning | |
| US8965582B2 (en) | Inverse kinematics | |
| CN108582068B (en) | Method for performing virtual simulation on circuit breaker assembly robot | |
| CN112417755A (en) | Master-slave mode surgical robot track prediction control method | |
| Ogbemhe et al. | Towards achieving a fully intelligent robotic arc welding: a review | |
| CN103331756A (en) | Mechanical arm motion control method | |
| CN113319857B (en) | Mechanical arm force and position hybrid control method and device, electronic equipment and storage medium | |
| CN110154024B (en) | Assembly control method based on long-term and short-term memory neural network incremental model | |
| CN110850719A (en) | Spatial non-cooperative target parameter self-tuning tracking method based on reinforcement learning | |
| CN103472724A (en) | Real-time control dynamics modeling method for multi-freedom-degree parallel mechanism | |
| CN113687659B (en) | Optimal trajectory generation method and system based on digital twinning | |
| CN110095983B (en) | Mobile robot prediction tracking control method based on path parameterization | |
| CN111897224B (en) | Multi-agent formation control method based on actor-critic reinforcement learning and fuzzy logic | |
| CN111168682A (en) | Limited time convergence sliding mode control method of parallel robot combined with robust precise differentiator for conveying | |
| CN116079714A (en) | Six-axis mechanical arm track planning method based on B spline | |
| CN117021118B (en) | Dynamic compensation method for digital twin track error of parallel robot | |
| CN110744552A (en) | Flexible mechanical arm motion control method based on singular perturbation theory | |
| CN112008726B (en) | Composite finite time control method based on exoskeleton robot actuator | |
| CN107511830A (en) | A kind of series parallel robot in five degrees of freedom controller parameter adaptively adjusts implementation method | |
| CN116068901A (en) | Flexible connecting rod mechanical arm control method based on self-adaptive finite time disturbance observer | |
| CN111857054B (en) | Numerical control system motion trajectory control method based on neural network | |
| Fu et al. | Integrated sliding mode control with input restriction, output feedback and repetitive learning for space robot with flexible-base, flexible-link and flexible-joint | |
| CN114260896A (en) | Compliance force control method and system for cooperative robot | |
| Jankowski et al. | Inverse dynamics and feedforward controllers for high precision position/force tracking of flexible joint robots | |
| CN111152213B (en) | Mechanical arm vibration compensation method and device based on hybrid control |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |