CN111639749A - Industrial robot friction force identification method based on deep learning - Google Patents
Industrial robot friction force identification method based on deep learning Download PDFInfo
- Publication number
- CN111639749A CN111639749A CN202010446594.4A CN202010446594A CN111639749A CN 111639749 A CN111639749 A CN 111639749A CN 202010446594 A CN202010446594 A CN 202010446594A CN 111639749 A CN111639749 A CN 111639749A
- Authority
- CN
- China
- Prior art keywords
- friction force
- torque
- moment
- deep learning
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 15
- 238000013135 deep learning Methods 0.000 title claims abstract description 12
- 230000036544 posture Effects 0.000 claims abstract description 4
- 238000003062 neural network model Methods 0.000 claims description 10
- 230000004913 activation Effects 0.000 claims description 3
- 238000013528 artificial neural network Methods 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims description 3
- 238000013136 deep learning model Methods 0.000 claims description 3
- 238000011478 gradient descent method Methods 0.000 claims description 3
- 238000013473 artificial intelligence Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/14—Force analysis or force optimisation, e.g. static or dynamic forces
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Data Mining & Analysis (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Computational Linguistics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Feedback Control In General (AREA)
Abstract
The invention provides an industrial robot friction force identification method based on deep learning, which comprises the following specific operation steps: enabling each axis of the robot to move independently in different postures, collecting position planning of each axis of the robot every 4ms, actually measuring motion track data, speed planning, calculating moment and actually measuring moment; subtracting the calculated torque from the measured torque to calculate an error torque; the invention provides an industrial robot friction force identification method based on deep learning, and solves the problems that an existing friction force model is inaccurate and an identification method is not accurate enough. The invention takes the error between the measured torque and the calculated torque as the friction force, and utilizes deep learning to identify the error torque, thereby solving the problem that the existing friction force model is not accurate enough. And obtaining a calculated torque which is more consistent with the actual friction force.
Description
Technical Field
The invention relates to the field of industrial robots, in particular to an industrial robot friction force identification method based on deep learning.
Background
The robot is a multi-joint manipulator or multi-freedom-degree machine device oriented to the industrial field, can automatically execute work, and is a machine which realizes various functions by means of self power and control capacity. The robot can accept human command and operate according to a preset program, and modern industrial robots can also perform actions according to a principle formulated by artificial intelligence technology.
Nowadays, through a robot dynamics model, the moments required during the robot movement can be calculated, which are usually the calculated moments that the control system can provide. In the control process of the industrial robot based on the model, the error between the calculated moment and the actually measured moment is required to be as small as possible, so that the control precision and stability can be better realized. When the friction force is modeled by the existing linear model and the nonlinear model, the friction force cannot be fitted accurately. The error between the actually measured torque and the calculated torque is used as the friction force, and the error torque is identified by utilizing deep learning, so that the problem that the existing friction force model is not accurate enough is solved. And obtaining a calculated torque which is more consistent with the actual friction force.
Disclosure of Invention
According to the technical problem, the invention provides an industrial robot friction force identification method based on deep learning, which comprises the following specific operation steps:
1) enabling each axis of the robot to move independently in different postures, collecting position planning of each axis of the robot every 4ms, actually measuring motion track data, speed planning, calculating moment and actually measuring moment;
2) subtracting the calculated torque from the measured torque to calculate an error torque;
3) building a deep learning model, wherein the neural network model with a feedforward structure is adopted, the built neural network model comprises a plurality of stacked structure modules, the number of the modules can be changed through API (application programming interface), and each stacked structure module comprises a batch standardization layer, a full connection layer and an activation layer;
4) inputting the position planning, the speed planning and the moment calculation of all the axes into a neural network model;
5) and eliminating a part which is too small and causes distortion in the original data when the relative error is calculated, wherein the relative error calculating method comprises the following steps: (calculated moment-measured moment)/measured moment;
6) 80% of collected data is divided into a training set, 20% of the collected data is divided into a testing set, a gradient descent learning friction model is adopted, and a neural network prediction model is obtained by utilizing a gradient descent method.
The invention has the beneficial effects that: the invention provides an industrial robot friction force identification method based on deep learning, and solves the problems that an existing friction force model is inaccurate and an identification method is not accurate enough. The invention takes the error between the measured torque and the calculated torque as the friction force, and utilizes deep learning to identify the error torque, thereby solving the problem that the existing friction force model is not accurate enough. And obtaining a calculated torque which is more consistent with the actual friction force.
Drawings
FIG. 1 is a schematic flow chart of the present invention.
FIG. 2 is a diagram of a deep learning neural network model structure according to the present invention.
Detailed Description
Example 1
When the method is used for identifying the friction force:
1) enabling each axis of the robot to move independently in different postures, collecting position planning of each axis of the robot every 4ms, actually measuring motion track data, speed planning, calculating moment and actually measuring moment;
2) subtracting the calculated torque from the measured torque to calculate an error torque;
3) building a deep learning model, wherein the neural network model with a feedforward structure is adopted, the built neural network model comprises a plurality of stacked structure modules, the number of the modules can be changed through API (application programming interface), and each stacked structure module comprises a batch standardization layer, a full connection layer and an activation layer;
4) inputting the position planning, the speed planning and the moment calculation of all the axes into a neural network model;
5) and eliminating a part which is too small and causes distortion in the original data when the relative error is calculated, wherein the relative error calculating method comprises the following steps: (calculated moment-measured moment)/measured moment;
6) 80% of collected data is divided into a training set, 20% of the collected data is divided into a testing set, a gradient descent learning friction model is adopted, and a neural network prediction model is obtained by utilizing a gradient descent method.
The foregoing shows and describes the general principles and broad features of the present invention and advantages thereof. While the invention has been described with respect to the above embodiments, it will be understood by those skilled in the art that the invention is not limited to the above embodiments, which are described in the specification and illustrated only to illustrate the principles of the invention, but that various changes and modifications may be made without departing from the spirit and scope of the invention as defined by the appended claims. The scope of the invention is defined by the appended claims and equivalents thereof.
Claims (1)
1. A friction force identification method of an industrial robot based on deep learning comprises the following specific operation steps:
1) enabling each axis of the robot to move independently in different postures, collecting position planning of each axis of the robot every 4ms, actually measuring motion track data, speed planning, calculating moment and actually measuring moment;
2) subtracting the calculated torque from the measured torque to calculate an error torque;
3) building a deep learning model, wherein the neural network model with a feedforward structure is adopted, the built neural network model comprises a plurality of stacked structure modules, the number of the modules can be changed through API (application programming interface), and each stacked structure module comprises a batch standardization layer, a full connection layer and an activation layer;
4) inputting the position planning, the speed planning and the moment calculation of all the axes into a neural network model;
5) and eliminating a part which is too small and causes distortion in the original data when the relative error is calculated, wherein the relative error calculating method comprises the following steps: (calculated moment-measured moment)/measured moment;
6) 80% of collected data is divided into a training set, 20% of the collected data is divided into a testing set, a gradient descent learning friction model is adopted, and a neural network prediction model is obtained by utilizing a gradient descent method.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010446594.4A CN111639749A (en) | 2020-05-25 | 2020-05-25 | Industrial robot friction force identification method based on deep learning |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010446594.4A CN111639749A (en) | 2020-05-25 | 2020-05-25 | Industrial robot friction force identification method based on deep learning |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111639749A true CN111639749A (en) | 2020-09-08 |
Family
ID=72330873
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010446594.4A Pending CN111639749A (en) | 2020-05-25 | 2020-05-25 | Industrial robot friction force identification method based on deep learning |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111639749A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112326187A (en) * | 2020-09-15 | 2021-02-05 | 南京航空航天大学 | Method for measuring friction force field by applying liquid crystal coating and deep learning algorithm |
| CN112326186A (en) * | 2020-09-15 | 2021-02-05 | 南京航空航天大学 | Liquid crystal coating color calibration method suitable for any irradiation and observation direction |
| CN114474078A (en) * | 2022-04-12 | 2022-05-13 | 季华实验室 | Friction force compensation method and device for mechanical arm, electronic equipment and storage medium |
| CN114594757A (en) * | 2020-12-07 | 2022-06-07 | 山东新松工业软件研究院股份有限公司 | Visual path planning method for cooperative robot |
| CN114670210A (en) * | 2022-05-30 | 2022-06-28 | 季华实验室 | Data acquisition method, pre-training method, device, electronic equipment and storage medium |
Citations (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050113973A1 (en) * | 2003-08-25 | 2005-05-26 | Sony Corporation | Robot and attitude control method of robot |
| CN101648264A (en) * | 2009-09-15 | 2010-02-17 | 邢台钢铁有限责任公司 | Method for distributing and controlling tension leveler torque in bloom continuous casting dynamic soft reduction process |
| US20110098860A1 (en) * | 2009-10-28 | 2011-04-28 | Honda Motor Co., Ltd. | Control device for legged mobile robot |
| CN103279037A (en) * | 2013-05-24 | 2013-09-04 | 华南理工大学 | Method for performing force follow-up motion control on robot on the basis of six-dimensional force/torque transducer |
| CN104638999A (en) * | 2014-08-27 | 2015-05-20 | 北京理工大学 | Segment neural network friction model based dual-motor servo system control method |
| CN106737657A (en) * | 2015-11-20 | 2017-05-31 | 沈阳新松机器人自动化股份有限公司 | A kind of robot security's control method and system based on dynamic system |
| CN107229334A (en) * | 2017-05-27 | 2017-10-03 | 东南大学 | It is a kind of based on active and passive device small-sized haptic feedback method |
| CN107498562A (en) * | 2017-04-21 | 2017-12-22 | 浙江工业大学 | Sixdegree-of-freedom simulation kinetic model discrimination method |
| CN107515530A (en) * | 2017-07-12 | 2017-12-26 | 北京控制工程研究所 | A kind of Nonlinear Control Allocation method based on depth autoencoder network |
| CN107703756A (en) * | 2017-11-03 | 2018-02-16 | 广州视源电子科技股份有限公司 | Kinetic parameters discrimination method, device, computer equipment and storage medium |
| CN108189036A (en) * | 2018-01-17 | 2018-06-22 | 广州视源电子科技股份有限公司 | Torque Control method, apparatus, robot and storage medium |
| CN108393892A (en) * | 2018-03-05 | 2018-08-14 | 厦门大学 | A kind of robot feedforward torque compensation method |
| CN108717492A (en) * | 2018-05-18 | 2018-10-30 | 浙江工业大学 | Manipulator Dynamic discrimination method based on improved artificial bee colony algorithm |
| CN108748144A (en) * | 2018-05-28 | 2018-11-06 | 上海优尼斯工业服务有限公司 | A kind of collision recognition method of man-machine collaboration mechanical arm |
| CN109702745A (en) * | 2019-01-18 | 2019-05-03 | 华南理工大学 | A kind of modeling method of joint of robot fluctuation moment of friction |
| CN109732605A (en) * | 2019-01-21 | 2019-05-10 | 厦门大学 | A kind of compensation method and system of joint of robot moment of friction |
| CN110059377A (en) * | 2019-04-02 | 2019-07-26 | 西南交通大学 | A kind of fuel battery service life prediction technique based on depth convolutional neural networks |
| CN110598862A (en) * | 2019-09-12 | 2019-12-20 | 山东浪潮人工智能研究院有限公司 | CNN model, training method thereof, terminal and computer-readable storage medium |
| CN110919633A (en) * | 2019-12-06 | 2020-03-27 | 泉州市微柏工业机器人研究院有限公司 | Robot position deviation method based on torque control |
| CN111177941A (en) * | 2020-01-03 | 2020-05-19 | 成都卡诺普自动化控制技术有限公司 | Robot friction force identification method |
-
2020
- 2020-05-25 CN CN202010446594.4A patent/CN111639749A/en active Pending
Patent Citations (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20050113973A1 (en) * | 2003-08-25 | 2005-05-26 | Sony Corporation | Robot and attitude control method of robot |
| CN101648264A (en) * | 2009-09-15 | 2010-02-17 | 邢台钢铁有限责任公司 | Method for distributing and controlling tension leveler torque in bloom continuous casting dynamic soft reduction process |
| US20110098860A1 (en) * | 2009-10-28 | 2011-04-28 | Honda Motor Co., Ltd. | Control device for legged mobile robot |
| CN103279037A (en) * | 2013-05-24 | 2013-09-04 | 华南理工大学 | Method for performing force follow-up motion control on robot on the basis of six-dimensional force/torque transducer |
| CN104638999A (en) * | 2014-08-27 | 2015-05-20 | 北京理工大学 | Segment neural network friction model based dual-motor servo system control method |
| CN106737657A (en) * | 2015-11-20 | 2017-05-31 | 沈阳新松机器人自动化股份有限公司 | A kind of robot security's control method and system based on dynamic system |
| CN107498562A (en) * | 2017-04-21 | 2017-12-22 | 浙江工业大学 | Sixdegree-of-freedom simulation kinetic model discrimination method |
| CN107229334A (en) * | 2017-05-27 | 2017-10-03 | 东南大学 | It is a kind of based on active and passive device small-sized haptic feedback method |
| CN107515530A (en) * | 2017-07-12 | 2017-12-26 | 北京控制工程研究所 | A kind of Nonlinear Control Allocation method based on depth autoencoder network |
| CN107703756A (en) * | 2017-11-03 | 2018-02-16 | 广州视源电子科技股份有限公司 | Kinetic parameters discrimination method, device, computer equipment and storage medium |
| CN108189036A (en) * | 2018-01-17 | 2018-06-22 | 广州视源电子科技股份有限公司 | Torque Control method, apparatus, robot and storage medium |
| CN108393892A (en) * | 2018-03-05 | 2018-08-14 | 厦门大学 | A kind of robot feedforward torque compensation method |
| CN108717492A (en) * | 2018-05-18 | 2018-10-30 | 浙江工业大学 | Manipulator Dynamic discrimination method based on improved artificial bee colony algorithm |
| CN108748144A (en) * | 2018-05-28 | 2018-11-06 | 上海优尼斯工业服务有限公司 | A kind of collision recognition method of man-machine collaboration mechanical arm |
| CN109702745A (en) * | 2019-01-18 | 2019-05-03 | 华南理工大学 | A kind of modeling method of joint of robot fluctuation moment of friction |
| CN109732605A (en) * | 2019-01-21 | 2019-05-10 | 厦门大学 | A kind of compensation method and system of joint of robot moment of friction |
| CN110059377A (en) * | 2019-04-02 | 2019-07-26 | 西南交通大学 | A kind of fuel battery service life prediction technique based on depth convolutional neural networks |
| CN110598862A (en) * | 2019-09-12 | 2019-12-20 | 山东浪潮人工智能研究院有限公司 | CNN model, training method thereof, terminal and computer-readable storage medium |
| CN110919633A (en) * | 2019-12-06 | 2020-03-27 | 泉州市微柏工业机器人研究院有限公司 | Robot position deviation method based on torque control |
| CN111177941A (en) * | 2020-01-03 | 2020-05-19 | 成都卡诺普自动化控制技术有限公司 | Robot friction force identification method |
Non-Patent Citations (3)
| Title |
|---|
| 中国地震局监测预报司编: "《地下流体数字观测技术》", 30 June 2002 * |
| 涂骁: "《基于动力学前馈的工业机器人运动控制关键技术研究》", 《中国优秀硕士学位论文全文数据库 (信息科技辑)》 * |
| 魏楠哲: "《空间机械臂柔性关节高精度控制研究》", 《中国博士学位论文全文数据库 (信息科技辑)》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112326187A (en) * | 2020-09-15 | 2021-02-05 | 南京航空航天大学 | Method for measuring friction force field by applying liquid crystal coating and deep learning algorithm |
| CN112326186A (en) * | 2020-09-15 | 2021-02-05 | 南京航空航天大学 | Liquid crystal coating color calibration method suitable for any irradiation and observation direction |
| CN112326187B (en) * | 2020-09-15 | 2022-04-08 | 南京航空航天大学 | Method for measuring friction force field by applying liquid crystal coating and deep learning algorithm |
| CN114594757A (en) * | 2020-12-07 | 2022-06-07 | 山东新松工业软件研究院股份有限公司 | Visual path planning method for cooperative robot |
| CN114474078A (en) * | 2022-04-12 | 2022-05-13 | 季华实验室 | Friction force compensation method and device for mechanical arm, electronic equipment and storage medium |
| CN114474078B (en) * | 2022-04-12 | 2022-06-17 | 季华实验室 | Friction force compensation method and device for mechanical arm, electronic equipment and storage medium |
| CN114670210A (en) * | 2022-05-30 | 2022-06-28 | 季华实验室 | Data acquisition method, pre-training method, device, electronic equipment and storage medium |
| CN114670210B (en) * | 2022-05-30 | 2022-08-02 | 季华实验室 | Data acquisition method, pre-training method, device, electronic equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111639749A (en) | Industrial robot friction force identification method based on deep learning | |
| CN111496781B (en) | Mechanical arm modeling, controlling and monitoring integrated system driven by digital twin | |
| CN110238839B (en) | Multi-shaft-hole assembly control method for optimizing non-model robot by utilizing environment prediction | |
| CN110228067B (en) | Double-arm robot combined operation testing method based on laser tracker | |
| CN110103225B (en) | Data-driven method and device for controlling repeated motion of mechanical arm | |
| CN112276944A (en) | Man-machine cooperation system control method based on intention recognition | |
| CN111872936B (en) | Robot collision detection system and method based on neural network | |
| CN108445764B (en) | Active compliance control strategy of Stewart platform | |
| CN113910218B (en) | Robot calibration method and device based on kinematic and deep neural network fusion | |
| CN110083160A (en) | A kind of method for planning track of robot based on deep learning | |
| CN103192363A (en) | Device and method for controlling planar three-freedom-degree flexible hinge parallel robot | |
| CN114310954B (en) | Self-adaptive lifting control method and system for nursing robot | |
| EP3929682B1 (en) | Estimating a mechanical degradation of a machine | |
| CN112192614A (en) | Man-machine cooperation based shaft hole assembling method for nuclear operation and maintenance robot | |
| WO2020024074A1 (en) | Voice control system for operating machinery | |
| Karlsson et al. | Detection and control of contact force transients in robotic manipulation without a force sensor | |
| US20230311309A1 (en) | Method and system for determining optimized program parameters for a robot program | |
| CN109352649A (en) | A kind of method for controlling robot and system based on deep learning | |
| CN108562763A (en) | Automobile wheel speed sensor tests system | |
| CN117484545A (en) | Intelligent multifunctional manipulator test platform system and test method | |
| CN110434854B (en) | Redundant manipulator visual servo control method and device based on data driving | |
| Chu et al. | Classifier-based approximator for friction compensation in high accelerated positioning system | |
| CN110488608B (en) | Intelligent kinetic parameter identification method and module for driving and controlling integrated control system | |
| CN114609969B (en) | Numerical control machine tool track error compensation method based on cloud computing | |
| CN114239938A (en) | State-based energy digital twin body construction method |
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 | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200908 |