CN117086886B - Robot dynamic error prediction method and system based on mechanism data hybrid driving - Google Patents
Robot dynamic error prediction method and system based on mechanism data hybrid driving Download PDFInfo
- Publication number
- CN117086886B CN117086886B CN202311346486.XA CN202311346486A CN117086886B CN 117086886 B CN117086886 B CN 117086886B CN 202311346486 A CN202311346486 A CN 202311346486A CN 117086886 B CN117086886 B CN 117086886B
- Authority
- CN
- China
- Prior art keywords
- motion
- parameters
- robot
- speed
- joint space
- 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 52
- 230000007246 mechanism Effects 0.000 title claims abstract description 25
- 238000012549 training Methods 0.000 claims description 15
- 230000008569 process Effects 0.000 claims description 12
- 238000003860 storage Methods 0.000 claims description 9
- 238000000342 Monte Carlo simulation Methods 0.000 claims description 7
- 238000004140 cleaning Methods 0.000 claims description 6
- 238000012544 monitoring process Methods 0.000 claims description 6
- 238000000605 extraction Methods 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 abstract description 2
- 230000015654 memory Effects 0.000 description 21
- 230000006870 function Effects 0.000 description 10
- 238000004891 communication Methods 0.000 description 5
- 238000004590 computer program Methods 0.000 description 5
- 239000011159 matrix material Substances 0.000 description 5
- 230000001133 acceleration Effects 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 4
- 238000013528 artificial neural network Methods 0.000 description 3
- 230000008878 coupling Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000010586 diagram Methods 0.000 description 3
- 238000004519 manufacturing process Methods 0.000 description 3
- 238000003062 neural network model Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 239000003638 chemical reducing agent Substances 0.000 description 1
- 238000013016 damping Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 239000000284 extract Substances 0.000 description 1
- 230000005484 gravity Effects 0.000 description 1
- 238000012994 industrial processing Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 230000003068 static effect Effects 0.000 description 1
- 239000000126 substance Substances 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
- B25J9/1602—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
- B25J9/1653—Programme controls characterised by the control loop parameters identification, estimation, stiffness, accuracy, error analysis
-
- 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
- B25J9/1661—Programme controls characterised by programming, planning systems for manipulators characterised by task planning, object-oriented languages
Landscapes
- Engineering & Computer Science (AREA)
- Robotics (AREA)
- Mechanical Engineering (AREA)
- Automation & Control Theory (AREA)
- Physics & Mathematics (AREA)
- Artificial Intelligence (AREA)
- Evolutionary Computation (AREA)
- Fuzzy Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Feedback Control In General (AREA)
Abstract
The invention relates to the technical field of robots, and particularly provides a robot dynamic error prediction method and system based on mechanism data hybrid driving, wherein the method comprises the following steps: acquiring a target position indicated by a control instruction; generating motion control parameters and motion parameters based on the target location, the motion parameters including location data; predicting joint space position residual errors based on motion control parameters and motion parameters and historical motion control parameters and motion parameters by utilizing a pre-trained LSTM model; generating a predicted position based on the motion parameter and the joint spatial position residual error; and outputting a difference value between the predicted position and the target position as a dynamic error. The invention greatly reduces the calculation amount of dynamic error prediction and improves the efficiency of dynamic error prediction.
Description
Technical Field
The invention belongs to the technical field of robots, and particularly relates to a robot dynamic error prediction method and system based on mechanism data hybrid driving.
Background
The industrial robot is a multi-joint manipulator or a multi-degree-of-freedom machine device widely used in the industrial field, has certain automaticity, and can realize various industrial processing and manufacturing functions by means of self power energy and control capability. Industrial robots are widely used in various industrial fields such as electronics, logistics, chemical industry, and the like. Industrial robots have high flexibility and low cost and are therefore widely used in manufacturing, however, the low tracking accuracy limits their application in high precision manufacturing.
The current method for improving the tracking precision of the industrial robot is to apply a robot dynamics model in the forward direction and to carry out dynamics simulation by means of Simscape, and the control mode realizes continuous control of a robot terminal executing mechanism in a working space. It requires movement strictly following a predetermined trajectory and speed within a certain precision range. And the speed is controllable, the movement track is smooth, and the task is completed. In this way, the dynamic error of the robot terminal actuator can be predicted.
However, the existing dynamic model has a complex physical process, is difficult to describe by using a mathematical model in an analytic form, and has the problems of strong nonlinear fitting capability, weak capability and the like.
Disclosure of Invention
Aiming at the problems of difficult analysis and untimely dynamic error prediction caused by weak strong nonlinear fitting capability in the prior art, the invention provides a robot dynamic error prediction method and system based on mechanism data hybrid driving, so as to solve the technical problems.
In a first aspect, the present invention provides a method for predicting dynamic errors of a robot based on a hybrid driving of mechanism data, including:
acquiring a target position indicated by a control instruction;
generating motion control parameters and motion parameters based on the target location, the motion parameters including location data;
predicting joint space position residual errors based on motion control parameters and motion parameters and historical motion control parameters and motion parameters by utilizing a pre-trained LSTM model;
generating a predicted position based on the motion parameter and the joint spatial position residual error;
outputting a difference between the predicted position and the target position as a dynamic error;
the training method of the LSTM model comprises the following steps:
enabling the tail end of the robot to run a random contour by using a Monte Carlo method, so that the motion trail of the robot fills the working space;
in the moving process of the robot, collecting the position and the speed of a motor, the position and the speed of a joint space and the position and the speed of a Cartesian space, acquiring corresponding target positions and actual positions from a robot control system, and storing all data into a data set;
training a pre-constructed LSTM model by utilizing the data set.
In an alternative embodiment, motion control parameters and motion parameters are generated based on the target location, the motion parameters including location data, including:
generating motion control parameters based on the target position and the current position of the robot by using a PID model, wherein the motion control parameters comprise motor position, speed and moment;
inputting the moment into a flexible dynamics model to obtain motion parameters, wherein the motion parameters comprise the position and the speed in a joint space and the position and the speed in a Cartesian space;
the motor position, speed and motion parameter are stored in an input data set after the motor position, speed and motion parameter are marked for generating time;
and monitoring the generation time of the data in the input data set, and clearing the corresponding data if the duration of the current moment of the generation time interval exceeds a set time threshold.
In an alternative embodiment, predicting joint spatial position residuals based on motion control parameters and motion parameters and historical motion control parameters and motion parameters using a pre-trained LSTM model includes:
and importing the input data set serving as input parameters into a pre-trained LSTM model to obtain a joint space position residual error predicted by the LSTM model.
In an alternative embodiment, generating the predicted position based on the motion parameter and the joint spatial position residual comprises:
extracting joint space positions from the motion parameters;
and outputting the sum of the joint space position and the joint space position residual error as a predicted position.
In a second aspect, the present invention provides a robot dynamic error prediction system based on a mechanism data hybrid drive, comprising:
the target acquisition module is used for acquiring a target position indicated by the control instruction;
the parameter acquisition module is used for generating motion control parameters and motion parameters based on the target position, wherein the motion parameters comprise position data;
the residual prediction module is used for predicting joint space position residual errors based on the motion control parameters and the motion parameters and the historical motion control parameters and the motion parameters by utilizing a pre-trained LSTM model;
the position prediction module is used for generating a predicted position based on the motion parameter and the joint space position residual error;
the error prediction module is used for outputting a difference value between the predicted position and the target position as a dynamic error;
the training method of the LSTM model comprises the following steps:
enabling the tail end of the robot to run a random contour by using a Monte Carlo method, so that the motion trail of the robot fills the working space;
in the moving process of the robot, collecting the position and the speed of a motor, the position and the speed of a joint space and the position and the speed of a Cartesian space, acquiring corresponding target positions and actual positions from a robot control system, and storing all data into a data set;
training a pre-constructed LSTM model by utilizing the data set.
In an alternative embodiment, the parameter acquisition module includes:
a first generation unit for generating motion control parameters based on the target position and the current position of the robot by using a PID model, wherein the motion control parameters comprise motor position, speed and moment;
the second generation unit is used for inputting the moment into the flexible dynamic model to obtain motion parameters, wherein the motion parameters comprise the position and the speed in the joint space and the position and the speed in the Cartesian space;
the parameter storage unit is used for storing the motor position, speed and motion parameter mark generation time into an input data set;
and the regular cleaning unit is used for monitoring the generation time of the data in the input data set, and cleaning the corresponding data if the duration of the current moment of the generation time interval exceeds a set time threshold.
In an alternative embodiment, the residual prediction module includes:
and the data input unit is used for importing the input data set serving as an input parameter into a pre-trained LSTM model to obtain a joint space position residual error predicted by the LSTM model.
In an alternative embodiment, the location prediction module includes:
a parameter extraction unit for extracting joint space position from the motion parameters;
and the residual error compensation unit is used for outputting the sum of the joint space position and the joint space position residual error as a predicted position.
In a third aspect, a terminal is provided, including:
a processor, a memory, wherein,
the memory is used for storing a computer program,
the processor is configured to call and run the computer program from the memory, so that the terminal performs the method of the terminal as described above.
In a fourth aspect, there is provided a computer storage medium having instructions stored therein which, when run on a computer, cause the computer to perform the method of the above aspects.
The robot dynamic error prediction method and system based on the mechanism data hybrid drive have the advantages that the motion control parameters of the servo motor are tracked and predicted through the PID model to obtain a joint space position, the LSTM model is utilized to conduct error prediction on the joint space position, further a predicted actual position is obtained, the predicted actual position is compared with the target position to obtain the dynamic error, the calculated amount of the dynamic error prediction is greatly reduced, and the efficiency of the dynamic error prediction is improved.
In addition, the invention has reliable design principle, simple structure and very wide application prospect.
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 required to be used in the description of the embodiments or the prior art will be briefly described below, and it will be obvious to those skilled in the art that other drawings can be obtained from these drawings without inventive effort.
FIG. 1 is a schematic flow chart of a method of one embodiment of the invention.
Fig. 2 is another schematic flow chart of a method of one embodiment of the invention.
FIG. 3 is a schematic block diagram of a system of one embodiment of the present invention.
Fig. 4 is a schematic structural diagram of a terminal according to an embodiment of the present invention.
Detailed Description
In order to make the technical solution of the present invention better understood by those skilled in the art, the technical solution of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention, and it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the present invention without making any inventive effort, shall fall within the scope of the present invention.
Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention.
The following explains key terms appearing in the present invention.
PID is: pro-port, integral, differential abbreviations. PID is a classical closed-loop control algorithm, and has the advantages of simple principle, easy realization, wide application range, mutually independent control parameters, simple parameter selection and the like. PID algorithms can be divided into two major categories, position PID and incremental PID. In practical programming applications, it is desirable to use a discretized PID algorithm to suit the environment in which the computer is used. The PID algorithm can control the motor speed, position and torque according to the current position and the target position.
A Long Short-Term Memory network (LSTM) is a time-loop neural network, which is specifically designed to solve the Long-Term dependency problem of a general RNN (loop neural network), and all RNNs have a chain form of a repeating neural network module. In a standard RNN, this repeated structural module has only a very simple structure, such as a tanh layer.
The robot dynamic error prediction method based on the mechanism data hybrid drive provided by the embodiment of the invention is executed by computer equipment, and correspondingly, the robot dynamic error prediction system based on the mechanism data hybrid drive operates in the computer equipment.
FIG. 1 is a schematic flow chart of a method of one embodiment of the invention. The execution subject of fig. 1 may be a robot dynamic error prediction system based on a mechanism data hybrid drive. The order of the steps in the flow chart may be changed and some may be omitted according to different needs.
As shown in fig. 1, the method includes:
step 110, obtaining a target position indicated by a control instruction;
step 120, generating motion control parameters and motion parameters based on the target position, the motion parameters including position data;
step 130, predicting joint space position residual errors based on motion control parameters and motion parameters and historical motion control parameters and motion parameters by utilizing a pre-trained LSTM model;
step 140, generating a predicted position based on the motion parameter and the joint spatial position residual error;
and step 150, outputting the difference between the predicted position and the target position as a dynamic error.
The training method of the LSTM model comprises the following steps:
enabling the tail end of the robot to run a random contour by using a Monte Carlo method, so that the motion trail of the robot fills the working space;
in the moving process of the robot, collecting the position and the speed of a motor, the position and the speed of a joint space and the position and the speed of a Cartesian space, acquiring corresponding target positions and actual positions from a robot control system, and storing all data into a data set;
training a pre-constructed LSTM model by utilizing the data set.
In order to facilitate understanding of the present invention, the principle of the robot dynamic error prediction method based on the hybrid driving of the mechanism data according to the present invention is used to further describe the robot dynamic error prediction method based on the hybrid driving of the mechanism data according to the present invention in combination with the process of predicting the robot dynamic error in the embodiment.
Specifically, referring to fig. 2, the method for predicting the dynamic error of the robot based on the hybrid driving of the mechanism data includes:
s1, acquiring a target position indicated by a control instruction.
The target position (coordinates) indicated by the instruction to be executed is extracted from the robot control system.
S2, generating motion control parameters and motion parameters based on the target position, wherein the motion parameters comprise position data.
Generating motion control parameters based on the target position and the current position of the robot by using a PID model, wherein the motion control parameters comprise motor position, speed and moment; inputting the moment into a flexible dynamics model to obtain motion parameters, wherein the motion parameters comprise the position and the speed in a joint space and the position and the speed in a Cartesian space; the motor position, speed and motion parameter are stored in an input data set after the motor position, speed and motion parameter are marked for generating time; and monitoring the generation time of the data in the input data set, and clearing the corresponding data if the duration of the current moment of the generation time interval exceeds a set time threshold.
Specifically, a mechanism model consisting of a PID model and a dynamics model is constructed. The PID model generates servo motor control parameters, i.e., motion control parameters, including motor position, speed, and torque, based on the target position and the current position. And inputting the moment into the dynamic model for analysis, and obtaining the motion parameters, namely the position and the speed in the joint space and the position and the speed in the Cartesian space. In order to improve the model accuracy, the dynamic model in the embodiment adopts a flexible dynamic model.
The flexible dynamic model is constructed, and when joint flexibility is considered, the recurrence relation of each connecting rod is as follows: link angular velocity, link angular acceleration, link linear velocity, link linear acceleration, and link centroid velocity. When joint flexibility is considered, the yaw rate of each link can be expressed as: link angular velocity, link centroid angular velocity. The equation of each degree of freedom obtained by using the Kane method is directed to the rotation angle of the connecting rod after the spring is denatured, and is directed to the degree of freedom of motor output: generalized inertial force, generalized primary force. The Kane equation can be used for establishing a dynamic model considering the joint flexibility system: f+f=0
The derived equation is 12 2 nd order differential equations, and when considering the influence of joint flexibility, the differential equation set is a time-varying differential equation, and numerical solution is needed. The form of the equation is known to be a time-varying nonlinear differential equation, wherein in a coefficient matrix, a mass matrix, a damping matrix and a rigidity matrix are all functions of coordinates, and a Newmark-beta method is used for solving, so that the equation has second-order precision.
In this embodiment, the constructed flexible dynamics model can be described as follows:
(1)
(2)
(3)
is a vector of joint position, velocity and acceleration, < >>Is a joint moment vector,/->Is an inertial matrix of the mass of the material,is centrifugal force and coriolis force, +.>Is the friction torque of the speed reducer, < >>Is gravity (or-> Is motor position, speed and acceleration, +.>Is the inertia of the motor rotor,/">Is the friction torque of the motor, ">Is a vector of the motor drive torque,/-, for example>Is a reduction ratio +.>Is torsional rigidity.
The dynamic model is the essence of the system and describes the physical process of the whole mechanical system, but the existing dynamic model still has the problems of simplifying the physical process, not modeling the part and the like. It is not practical to describe the robot dynamic error using only the dynamic model, and most researchers apply the dynamic model to moment prediction, and few researchers apply the dynamic error prediction.
And taking the motor position, speed and motion parameters within 30min as time sequence data required by prediction.
S3, predicting joint space position residual errors based on the motion control parameters and the motion parameters and the historical motion control parameters and the motion parameters by utilizing a pre-trained LSTM model.
Enabling the tail end of the robot to run a random contour by using a Monte Carlo method, so that the motion trail of the robot fills the working space; in the moving process of the robot, collecting the position and the speed of a motor, the position and the speed of a joint space and the position and the speed of a Cartesian space, acquiring corresponding target positions and actual positions from a robot control system, and storing all data into a data set; training a pre-constructed LSTM model by utilizing the data set. The training data generated by the method is richer and more comprehensive, and is beneficial to improving the prediction precision and generalization of the model. By introducing the calculation data of the mechanism model as the training data of the neural network model, the non-end-to-end data set form can fully integrate the mechanism model into the neural network model, so that the neural network model extracts more characteristic elements, and the model prediction precision and stability are improved.
And (2) taking the input data set in the step (S2) as an input parameter to be imported into a pre-trained LSTM model to obtain a joint space position residual error predicted by the LSTM model.
S4, generating a predicted position based on the motion parameter and the joint space position residual error.
Extracting joint space positions from the motion parameters; and outputting the sum of the joint space position and the joint space position residual error as a predicted position.
Specifically, predicted position = joint spatial position + joint spatial position residual.
That is, the joint space position output by the mechanism model is compensated by using the position residual error predicted by the LSTM model, and the predicted actual position is obtained.
S5, outputting a difference value between the predicted position and the target position as a dynamic error.
And (3) performing difference between the predicted position obtained in the step (S4) and the target position indicated by the control instruction to obtain a dynamic error.
The dynamic error of the robot can be obtained in real time by the method, and the dynamic error prediction efficiency of the robot is improved.
In some embodiments, the system 300 may include a plurality of functional modules consisting of computer program segments. The computer program of each program segment in the system 300 for predicting the dynamic error of the robot based on the hybrid driving of the mechanism data may be stored in a memory of a computer device and executed by at least one processor to perform (see fig. 1 for details) the functions of the prediction of the dynamic error of the robot based on the hybrid driving of the mechanism data.
In this embodiment, the robot dynamic error prediction system 300 based on the hybrid driving of the mechanism data may be divided into a plurality of functional modules according to the functions performed by the system, as shown in fig. 3. The functional module may include: a target acquisition module 310, a parameter acquisition module 320, a residual prediction module 330, a position prediction module 340, and an error prediction module 350. The module referred to in the present invention refers to a series of computer program segments capable of being executed by at least one processor and of performing a fixed function, stored in a memory. In the present embodiment, the functions of the respective modules will be described in detail in the following embodiments.
A target obtaining module 310, configured to obtain a target position indicated by the control instruction;
a parameter acquisition module 320 for generating motion control parameters and motion parameters based on the target location, the motion parameters including location data;
the residual prediction module 330 is configured to predict a joint spatial position residual based on the motion control parameter and the motion parameter and the historical motion control parameter and the motion parameter by using a pre-trained LSTM model;
a position prediction module 340 for generating a predicted position based on the motion parameter and the joint spatial position residual;
an error prediction module 350, configured to output a difference between the predicted position and the target position as a dynamic error;
the training method of the LSTM model comprises the following steps:
enabling the tail end of the robot to run a random contour by using a Monte Carlo method, so that the motion trail of the robot fills the working space;
in the moving process of the robot, collecting the position and the speed of a motor, the position and the speed of a joint space and the position and the speed of a Cartesian space, acquiring corresponding target positions and actual positions from a robot control system, and storing all data into a data set;
training a pre-constructed LSTM model by utilizing the data set.
Optionally, as an embodiment of the present invention, the parameter obtaining module includes:
a first generation unit for generating motion control parameters based on the target position and the current position of the robot by using a PID model, wherein the motion control parameters comprise motor position, speed and moment;
the second generation unit is used for inputting the moment into the flexible dynamic model to obtain motion parameters, wherein the motion parameters comprise the position and the speed in the joint space and the position and the speed in the Cartesian space;
the parameter storage unit is used for storing the motor position, speed and motion parameter mark generation time into an input data set;
and the regular cleaning unit is used for monitoring the generation time of the data in the input data set, and cleaning the corresponding data if the duration of the current moment of the generation time interval exceeds a set time threshold.
Optionally, as an embodiment of the present invention, the residual prediction module includes:
and the data input unit is used for importing the input data set serving as an input parameter into a pre-trained LSTM model to obtain a joint space position residual error predicted by the LSTM model.
Optionally, as an embodiment of the present invention, the location prediction module includes:
a parameter extraction unit for extracting joint space position from the motion parameters;
and the residual error compensation unit is used for outputting the sum of the joint space position and the joint space position residual error as a predicted position.
Fig. 4 is a schematic structural diagram of a terminal 400 according to an embodiment of the present invention, where the terminal 400 may be used to execute the method for predicting dynamic errors of a robot based on hybrid driving of mechanism data according to the embodiment of the present invention.
The terminal 400 may include: processor 410, memory 420, and communication module 430. The components may communicate via one or more buses, and it will be appreciated by those skilled in the art that the configuration of the server as shown in the drawings is not limiting of the invention, as it may be a bus-like structure, a star-like structure, or include more or fewer components than shown, or may be a combination of certain components or a different arrangement of components.
The memory 420 may be used to store instructions for execution by the processor 410, and the memory 420 may be implemented by any type of volatile or nonvolatile memory terminal or combination thereof, such as Static Random Access Memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk, or optical disk. The execution of the instructions in memory 420, when executed by processor 410, enables terminal 400 to perform some or all of the steps in the method embodiments described below.
The processor 410 is a control center of the storage terminal, connects various parts of the entire electronic terminal using various interfaces and lines, and performs various functions of the electronic terminal and/or processes data by running or executing software programs and/or modules stored in the memory 420, and invoking data stored in the memory. The processor may be comprised of an integrated circuit (Integrated Circuit, simply referred to as an IC), for example, a single packaged IC, or may be comprised of a plurality of packaged ICs connected to the same function or different functions. For example, the processor 410 may include only a central processing unit (Central Processing Unit, simply CPU). In the embodiment of the invention, the CPU can be a single operation core or can comprise multiple operation cores.
And a communication module 430, configured to establish a communication channel, so that the storage terminal can communicate with other terminals. Receiving user data sent by other terminals or sending the user data to other terminals.
The present invention also provides a computer storage medium in which a program may be stored, which program may include some or all of the steps in the embodiments provided by the present invention when executed. The storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM), a random-access memory (random access memory, RAM), or the like.
Therefore, the motion control parameters of the servo motor are tracked and predicted by the PID model to obtain a joint space position, the LSTM model is utilized to conduct error prediction on the joint space position to obtain a predicted actual position, the predicted actual position is compared with the target position to obtain a dynamic error, the calculated amount of the dynamic error prediction is greatly reduced, the efficiency of the dynamic error prediction is improved, and the technical effects achieved by the embodiment can be seen from the description above and are not repeated.
It will be apparent to those skilled in the art that the techniques of embodiments of the present invention may be implemented in software plus a necessary general purpose hardware platform. Based on such understanding, the technical solution in the embodiments of the present invention may be embodied essentially or in a part contributing to the prior art in the form of a software product stored in a storage medium such as a U-disc, a mobile hard disc, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk or an optical disk, etc. various media capable of storing program codes, including several instructions for causing a computer terminal (which may be a personal computer, a server, or a second terminal, a network terminal, etc.) to execute all or part of the steps of the method described in the embodiments of the present invention.
The same or similar parts between the various embodiments in this specification are referred to each other. In particular, for the terminal embodiment, since it is substantially similar to the method embodiment, the description is relatively simple, and reference should be made to the description in the method embodiment for relevant points.
In the several embodiments provided by the present invention, it should be understood that the disclosed systems and methods may be implemented in other ways. For example, the system embodiments described above are merely illustrative, e.g., the division of the modules is merely a logical function division, and there may be additional divisions when actually implemented, e.g., multiple modules or components may be combined or integrated into another system, or some features may be omitted or not performed. Alternatively, the coupling or direct coupling or communication connection shown or discussed with respect to each other may be through some interface, indirect coupling or communication connection of systems or modules, electrical, mechanical, or other form.
The modules described as separate components may or may not be physically separate, and components shown as modules may or may not be physical modules, i.e., may be located in one place, or may be distributed over a plurality of network modules. Some or all of the modules may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional module in each embodiment of the present invention may be integrated into one processing module, or each module may exist alone physically, or two or more modules may be integrated into one module.
Although the present invention has been described in detail by way of preferred embodiments with reference to the accompanying drawings, the present invention is not limited thereto. Various equivalent modifications and substitutions may be made in the embodiments of the present invention by those skilled in the art without departing from the spirit and scope of the present invention, and it is intended that all such modifications and substitutions be within the scope of the present invention/be within the scope of the present invention as defined by the appended claims. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.
Claims (8)
1. The robot dynamic error prediction method based on the mechanism data hybrid driving is characterized by comprising the following steps of:
acquiring a target position indicated by a control instruction;
generating motion control parameters and motion parameters based on the target location, the motion parameters including location data;
predicting joint space position residual errors based on motion control parameters and motion parameters and historical motion control parameters and motion parameters by utilizing a pre-trained LSTM model;
generating a predicted position based on the motion parameter and the joint spatial position residual error;
outputting a difference between the predicted position and the target position as a dynamic error;
the training method of the LSTM model comprises the following steps:
enabling the tail end of the robot to run a random contour by using a Monte Carlo method, so that the motion trail of the robot fills the working space;
in the moving process of the robot, collecting the position and the speed of a motor, the position and the speed of a joint space and the position and the speed of a Cartesian space, acquiring corresponding target positions and actual positions from a robot control system, and storing all data into a data set;
training a pre-constructed LSTM model by utilizing the data set.
2. The method of claim 1, wherein generating motion control parameters and motion parameters based on the target location, the motion parameters including location data, comprises:
generating motion control parameters based on the target position and the current position of the robot by using a PID model, wherein the motion control parameters comprise motor position, speed and moment;
inputting the moment into a flexible dynamics model to obtain motion parameters, wherein the motion parameters comprise the position and the speed in a joint space and the position and the speed in a Cartesian space;
the motor position, speed and motion parameter are stored in an input data set after the motor position, speed and motion parameter are marked for generating time;
and monitoring the generation time of the data in the input data set, and clearing the corresponding data if the duration of the current moment of the generation time interval exceeds a set time threshold.
3. The method of claim 2, wherein predicting joint spatial position residuals based on motion control parameters and motion parameters and historical motion control parameters and motion parameters using a pre-trained LSTM model comprises:
and importing the input data set serving as input parameters into a pre-trained LSTM model to obtain a joint space position residual error predicted by the LSTM model.
4. The method of claim 1, wherein generating the predicted position based on the motion parameter and the joint spatial position residual comprises:
extracting joint space positions from the motion parameters;
and outputting the sum of the joint space position and the joint space position residual error as a predicted position.
5. The robot dynamic error prediction system based on the mechanism data hybrid driving is characterized by comprising:
the target acquisition module is used for acquiring a target position indicated by the control instruction;
the parameter acquisition module is used for generating motion control parameters and motion parameters based on the target position, wherein the motion parameters comprise position data;
the residual prediction module is used for predicting joint space position residual errors based on the motion control parameters and the motion parameters and the historical motion control parameters and the motion parameters by utilizing a pre-trained LSTM model;
the position prediction module is used for generating a predicted position based on the motion parameter and the joint space position residual error;
the error prediction module is used for outputting a difference value between the predicted position and the target position as a dynamic error;
the training method of the LSTM model comprises the following steps:
enabling the tail end of the robot to run a random contour by using a Monte Carlo method, so that the motion trail of the robot fills the working space;
in the moving process of the robot, collecting the position and the speed of a motor, the position and the speed of a joint space and the position and the speed of a Cartesian space, acquiring corresponding target positions and actual positions from a robot control system, and storing all data into a data set;
training a pre-constructed LSTM model by utilizing the data set.
6. The system of claim 5, wherein the parameter acquisition module comprises:
a first generation unit for generating motion control parameters based on the target position and the current position of the robot by using a PID model, wherein the motion control parameters comprise motor position, speed and moment;
the second generation unit is used for inputting the moment into the flexible dynamic model to obtain motion parameters, wherein the motion parameters comprise the position and the speed in the joint space and the position and the speed in the Cartesian space;
the parameter storage unit is used for storing the motor position, speed and motion parameter mark generation time into an input data set;
and the regular cleaning unit is used for monitoring the generation time of the data in the input data set, and cleaning the corresponding data if the duration of the current moment of the generation time interval exceeds a set time threshold.
7. The system of claim 6, wherein the residual prediction module comprises:
and the data input unit is used for importing the input data set serving as an input parameter into a pre-trained LSTM model to obtain a joint space position residual error predicted by the LSTM model.
8. The system of claim 5, wherein the location prediction module comprises:
a parameter extraction unit for extracting joint space position from the motion parameters;
and the position prediction unit is used for outputting the sum of the joint space position and the joint space position residual error as a predicted position.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311346486.XA CN117086886B (en) | 2023-10-18 | 2023-10-18 | Robot dynamic error prediction method and system based on mechanism data hybrid driving |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311346486.XA CN117086886B (en) | 2023-10-18 | 2023-10-18 | Robot dynamic error prediction method and system based on mechanism data hybrid driving |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117086886A CN117086886A (en) | 2023-11-21 |
| CN117086886B true CN117086886B (en) | 2023-12-22 |
Family
ID=88780576
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311346486.XA Active CN117086886B (en) | 2023-10-18 | 2023-10-18 | Robot dynamic error prediction method and system based on mechanism data hybrid driving |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117086886B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117272846A (en) * | 2023-11-22 | 2023-12-22 | 山东建筑大学 | Dynamic response prediction algorithm for two-degree-of-freedom rotary pitching motion mechanism |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021164276A1 (en) * | 2020-07-31 | 2021-08-26 | 平安科技（深圳）有限公司 | Target tracking method and apparatus, computer device, and storage medium |
| CN114131611A (en) * | 2021-12-16 | 2022-03-04 | 华中科技大学 | Joint error offline compensation method, system and terminal for robot gravity pose decomposition |
| CN114789454A (en) * | 2022-06-24 | 2022-07-26 | 浙江大学 | Robot digital twin track completion method based on LSTM and inverse kinematics |
| CN114995132A (en) * | 2022-05-26 | 2022-09-02 | 哈尔滨工业大学（深圳） | Multi-arm spacecraft model prediction control method, equipment and medium based on Gaussian mixture process |
| CN115284276A (en) * | 2022-03-28 | 2022-11-04 | 江南大学 | Robot joint torque control method based on long-term and short-term memory network |
| CN115577755A (en) * | 2022-11-28 | 2023-01-06 | 中环服(成都)科技有限公司 | Robot posture correction method, apparatus, computer device, and storage medium |
-
2023
- 2023-10-18 CN CN202311346486.XA patent/CN117086886B/en active Active
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2021164276A1 (en) * | 2020-07-31 | 2021-08-26 | 平安科技（深圳）有限公司 | Target tracking method and apparatus, computer device, and storage medium |
| CN114131611A (en) * | 2021-12-16 | 2022-03-04 | 华中科技大学 | Joint error offline compensation method, system and terminal for robot gravity pose decomposition |
| CN115284276A (en) * | 2022-03-28 | 2022-11-04 | 江南大学 | Robot joint torque control method based on long-term and short-term memory network |
| CN114995132A (en) * | 2022-05-26 | 2022-09-02 | 哈尔滨工业大学（深圳） | Multi-arm spacecraft model prediction control method, equipment and medium based on Gaussian mixture process |
| CN114789454A (en) * | 2022-06-24 | 2022-07-26 | 浙江大学 | Robot digital twin track completion method based on LSTM and inverse kinematics |
| CN115577755A (en) * | 2022-11-28 | 2023-01-06 | 中环服(成都)科技有限公司 | Robot posture correction method, apparatus, computer device, and storage medium |
Non-Patent Citations (1)
| Title |
|---|
| 周恒旭 ; 何志敏 ; 周燕 ; .基于长短期记忆网络的机械臂逆运动学解.机械与电子.2020,(06),全文. * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117086886A (en) | 2023-11-21 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Truong et al. | A backstepping global fast terminal sliding mode control for trajectory tracking control of industrial robotic manipulators | |
| Bhardwaj et al. | Storm: An integrated framework for fast joint-space model-predictive control for reactive manipulation | |
| EP2188685B1 (en) | Inverse kinematics | |
| CN117086886B (en) | Robot dynamic error prediction method and system based on mechanism data hybrid driving | |
| Vladareanu et al. | The navigation of mobile robots in non-stationary and non-structured environments | |
| CN111168682B (en) | Parallel robot and robust precise differentiator combined finite time convergence sliding mode control method | |
| Urrea et al. | Design and validation of a dynamic parameter identification model for industrial manipulator robots | |
| De La Cruz et al. | Online learning of inverse dynamics via gaussian process regression | |
| Axelsson | On Sensor Fusion Applied to Industrial Manipulators | |
| CN114967465A (en) | Trajectory planning method and device, electronic equipment and storage medium | |
| Jin et al. | Task-driven hybrid model reduction for dexterous manipulation | |
| Chaudhary et al. | Fractional order fast terminal sliding mode control scheme for tracking control of robot manipulators | |
| Wesselowski et al. | A dual-mode model predictive controller for robot formations | |
| US20230191604A1 (en) | Robot stability control method, robot and computer-readable storage medium | |
| Cifuentes-Molano et al. | Comparison of Different Control Techniques on a Bipedal Robot of 6 Degrees of Freedom. | |
| CN116079747A (en) | Robot cross-body control method, system, computer equipment and storage medium | |
| Nawrocka et al. | Neural network control for robot manipulator | |
| Vladareanu et al. | The petri nets and Markov chains approach for the walking robots dynamical stability control | |
| Urrea et al. | Design of a generalized dynamic model and a trajectory control and position strategy for n-link underactuated revolute planar robots | |
| Gattringer et al. | Extending continuous path trajectories to point-to-point trajectories by varying intermediate points | |
| Diachenko et al. | Control laws of electric drives as a result of an in-depth kinematic analysis of the delta robot | |
| CN112632803A (en) | Tracking control method and device, electronic equipment and storage medium | |
| Ji et al. | Weight lifting trajectory optimization for variable stiffness actuated robot | |
| Valera et al. | Solving the inverse dynamic control for low cost real-time industrial robot control applications | |
| Lämmle et al. | Process simulation using physical joining models for learning robot-based assembly tasks |
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 |