CN114167722A - Parallel robot tracking control method based on super-exponential convergence neural network - Google Patents
Parallel robot tracking control method based on super-exponential convergence neural network Download PDFInfo
- Publication number
- CN114167722A CN114167722A CN202111421980.9A CN202111421980A CN114167722A CN 114167722 A CN114167722 A CN 114167722A CN 202111421980 A CN202111421980 A CN 202111421980A CN 114167722 A CN114167722 A CN 114167722A
- Authority
- CN
- China
- Prior art keywords
- time
- matlab
- real
- parallel robot
- leg
- 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.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 25
- 238000013528 artificial neural network Methods 0.000 title claims abstract description 12
- 239000011159 matrix material Substances 0.000 claims abstract description 21
- 230000008569 process Effects 0.000 claims abstract description 13
- 239000012636 effector Substances 0.000 claims abstract description 12
- 230000001133 acceleration Effects 0.000 claims abstract description 8
- 238000005094 computer simulation Methods 0.000 claims abstract description 5
- 238000004891 communication Methods 0.000 claims abstract description 4
- 230000004913 activation Effects 0.000 claims description 2
- 230000006978 adaptation Effects 0.000 claims description 2
- 238000004422 calculation algorithm Methods 0.000 abstract description 9
- 238000004088 simulation Methods 0.000 abstract description 6
- 230000000977 initiatory effect Effects 0.000 description 2
- 238000013459 approach Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000002789 length control Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 239000003973 paint Substances 0.000 description 1
- 230000000306 recurrent effect Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000005507 spraying Methods 0.000 description 1
- 230000001360 synchronised effect Effects 0.000 description 1
- 230000009897 systematic effect Effects 0.000 description 1
- 238000003466 welding Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05B—CONTROL OR REGULATING SYSTEMS IN GENERAL; FUNCTIONAL ELEMENTS OF SUCH SYSTEMS; MONITORING OR TESTING ARRANGEMENTS FOR SUCH SYSTEMS OR ELEMENTS
- G05B13/00—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion
- G05B13/02—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric
- G05B13/04—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators
- G05B13/042—Adaptive control systems, i.e. systems automatically adjusting themselves to have a performance which is optimum according to some preassigned criterion electric involving the use of models or simulators in which a parameter or coefficient is automatically adjusted to optimise the performance
Landscapes
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Numerical Control (AREA)
- Feedback Control In General (AREA)
Abstract
The invention discloses a tracking control method of a parallel robot based on a super-exponential convergence neural network, which comprises the following specific steps: a user initiates a communication request at an MATLAB end, establishes connection and inputs an expected path and an expected speed; MATLAB reads the real-time position, speed and acceleration of the parallel robot end effector in VREP, and reads the real-time acceleration of each supporting leg; calculating by utilizing real-time data at an MATLAB end to obtain a real-time control signal and a coefficient matrix updating signal; calculating the MATLAB end control signal and the updating signal to obtain an updated coefficient matrix and control information of each supporting leg; and inputting the obtained leg control information into the VREP and executing a dynamic simulation process. The invention utilizes real-time state and data of the parallel robot with uncertainty, realizes accurate and rapid tracking and control of the parallel robot through an online algorithm, and dynamically simulates the experimental process through a VREP simulation platform.
Description
Technical Field
The invention relates to the field of parallel mechanical arm motion control algorithms, in particular to a method for realizing accurate tracking control of a parallel robot with uncertainty by utilizing an ZNN model and a super-exponential convergence factor.
Background
The parallel robot is a closed-loop mechanism which is driven in a parallel mode by connecting a movable platform and a fixed platform through more than two independent kinematic chains. The parallel robot has the excellent characteristics of no accumulated error, high precision, quick response, high bearing capacity and the like, and is widely applied to the industries of aviation, aerospace, seabed operation, underground mining and the like. In the case of the present invention, the stewart platform is a typical robot with six degrees of freedom.
The real-time tracking control of the parallel robot mainly controls the movement of an end effector installed on a mobile platform along a desired path in a robot working space through online calculation and guidance. The real-time control parallel robot has wide application in the work with intensive labor and high precision requirement, such as assembly, welding, paint spraying and the like.
Recurrent Neural Networks (RNNs) are often applied to solve real-time problems, while return-to-Zero Neural Networks (ZNN) can handle multiple state-dimensional problems and thus can be considered as systematic approaches to solving various time-varying problems, including time-varying tracking control problems. For the parallel robot with uncertainty, the real-time tracking control problem is solved by combining the super-exponential convergence factor with the return-to-zero neural network, and the SEC-ZNN has the advantages of both the super-exponential convergence factor and the return-to-zero neural network, so that the effect of accurate control is achieved while the robot is quickly converged.
Disclosure of Invention
Aiming at the defects of the prior art, the invention provides a tracking control method of a parallel robot based on a super-exponential convergence neural network, which has the characteristic of super-exponential convergence.
In order to solve the technical problems, the invention is realized by the following technical scheme:
a tracking control method of a parallel robot based on a super-exponential convergence neural network comprises the following steps:
1) a user initiates a communication request at an MATLAB end to establish connection;
3) MATLAB reads real-time position r of parallel robot end effector in VREPa(t), real time velocityAnd accelerationReading real-time acceleration of each leg
4) Calculating to obtain a real-time control signal at an MATLAB end by using the real-time data in the step 3) according to a kinetic equationSum coefficient matrix update signal
5) Calculating to obtain an updated coefficient matrix at the MATLAB terminal by using the control signal and the updating signal in the step 4)And each leg control information l (t + Δ t).
6) Inputting the leg control information l (t + delta t) obtained by the MATLAB terminal into VREP and executing a dynamic simulation process.
The dynamic model of the step 4) comprises the following specific methods:
wherein the formula (1) is a landing leg control signal iterative formula,the leg control signal is the control signal for the leg at time t,for the matrix of coefficients at the time t,for the desired velocity of the end effector at time t,is to set a parameter for controlling convergence in the tracking process, Ψ (-) is an activation function matrix and each element therein is an odd function that monotonically increases, rd(t) and ra(t) represents the desired path and the actual path of the end effector at time t, respectively.
Equation (2) is an iterative equation of the coefficient matrix, in whichThe coefficient matrix update signal is updated for time t,for the acceleration of each leg at time t,for the matrix of coefficients at the time t,andrespectively representing the actual velocity and the actual acceleration of the end effector at time t,is to set parameters for controlling convergence in the adaptation process, superscriptRepresenting a pseudo-inverse result of a vector or matrix.
The planning problem is solved by using an SEC-ZNN algorithm and real-time dynamic simulation is carried out at a VREP end, in the step 4), in an iteration process, an index factor exp (t) is used, so that a path is rapidly converged and is more accurately controlled, in the step 6), a VREP platform is used for real-time dynamic simulation, and the motion process of the robot is really shown.
The invention fully utilizes the real-time state and data of the parallel robot with uncertainty, realizes the accurate and rapid tracking and control of the parallel robot through the online algorithm, and dynamically simulates the experimental process through the VREP simulation platform, thereby providing more accurate reference for transferring the algorithm to the real object platform in the future.
Drawings
FIG. 1 is a flow chart of the present invention;
FIG. 2 is a flow chart of the SEC-ZNN algorithm;
FIG. 3 is the result of an operation on the simulation platform.
Detailed Description
The invention is further described with reference to the following detailed description and accompanying drawings:
FIG. 1 shows a tracking control method of a parallel robot based on a super-exponential convergence neural network, which comprises the following steps:
1) performing tracking control on the parallel robot with uncertainty by using an SEC-ZNN algorithm;
2) the MATLAB is used as a remote control end, the VREP platform is used as an actual simulation end, and the simulation parallel robot carries out tracking control in the actual motion process with uncertainty;
therefore, the process of the invention is as follows:
1) and initiating a connection establishment request at the MATLAB end, and initiating a VREP end to establish connection corresponding to the request.
2) And sending a control instruction by the MATLAB terminal to read and set related information such as the initial robot posture, the control joint handle, the communication time interval and the like of the VREP terminal, and generating a simplified robot model at the MATLAB terminal.
3) And when the iteration is not finished and the connection is not disconnected, calling a synchronous trigger and reading the dynamic parameters.
4) And calculating by using an SEC-ZNN algorithm to obtain the leg control information. The SEC-ZNN algorithm comprises the following specific contents:
according to the kinetic model:
and calculating the iteration quantity of the leg length and the coefficient matrix corresponding to the time t. For the obtained iteration quantity, according to an iteration formula:
and iteratively updating the leg length control quantity and the coefficient matrix.
5) The MATLAB end sends out a control signal, and the VREP end receives the signal and then utilizes the PID controller to dynamically simulate the leg movement.
Fig. 3 shows the tracking control result of a simulation experiment. The track in the figure is the motion track of the end effector, and the end effector is arranged on the upper surface of the Stewart parallel robot.
The above-described embodiments of the present invention do not limit the scope of the present invention. Any modification, equivalent replacement, and improvement made within the spirit and scope of the present invention shall be included in the protection scope of the claims of the present invention.
Claims (2)
1. A tracking control method of a parallel robot based on a super-exponential convergence neural network is characterized by comprising the following steps:
1) a user initiates a communication request at an MATLAB end to establish connection;
3) MATLAB reads real-time position r of parallel robot end effector in VREPa(t), real time velocityAnd accelerationReading real-time acceleration of each leg
4) Calculating to obtain a real-time control signal at an MATLAB end by using the real-time data in the step 3) according to a kinetic equationSum coefficient matrix update signal
The specific method for calculating according to the kinetic equation comprises the following steps:
wherein the formula (1) is a landing leg control signal iterative formula,the leg control signal is the control signal for the leg at time t,for the matrix of coefficients at the time t,for the desired velocity of the end effector at time t,is to set a parameter for controlling convergence in the tracking process, Ψ (-) is an activation function matrix and each element therein is an odd function that monotonically increases, rd(t) and ra(t) respectively representing the expected path and the actual path of the end effector at the time t, exp (t) being an exponential factor;
equation (2) is an iterative equation of the coefficient matrix, in whichThe coefficient matrix update signal is updated for time t,for the acceleration of each leg at time t,for the matrix of coefficients at the time t,andrespectively representing the actual velocity and the actual acceleration of the end effector at time t,is to set parameters for controlling convergence in the adaptation process, superscriptA pseudo-inverse matrix result representing a vector or matrix;
5) calculating to obtain an updated coefficient matrix at the MATLAB terminal by using the control signal and the updating signal in the step 4)And each leg control information iota (t + delta t);
6) inputting the leg control information l (t + delta t) obtained by the MATLAB terminal into VREP and executing a dynamic simulation process.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111421980.9A CN114167722B (en) | 2021-11-26 | 2021-11-26 | Parallel robot tracking control method based on super-exponential convergence neural network |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202111421980.9A CN114167722B (en) | 2021-11-26 | 2021-11-26 | Parallel robot tracking control method based on super-exponential convergence neural network |
Publications (2)
Publication Number | Publication Date |
---|---|
CN114167722A true CN114167722A (en) | 2022-03-11 |
CN114167722B CN114167722B (en) | 2024-03-29 |
Family
ID=80481126
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202111421980.9A Active CN114167722B (en) | 2021-11-26 | 2021-11-26 | Parallel robot tracking control method based on super-exponential convergence neural network |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114167722B (en) |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2018176854A1 (en) * | 2017-03-27 | 2018-10-04 | 华南理工大学 | Method for programming repeating motion of redundant robotic arm |
CN110134013A (en) * | 2019-05-08 | 2019-08-16 | 杭州电子科技大学 | The parallel mechanical arm finite time convergence control motion control method of external disturbance can be fought |
CN111443712A (en) * | 2020-03-30 | 2020-07-24 | 杭州电子科技大学 | Three-dimensional path planning method based on longicorn group search algorithm |
CN111872933A (en) * | 2019-11-25 | 2020-11-03 | 浙江大学宁波理工学院 | SCARA robot trajectory tracking control method based on improved quadratic iterative learning control |
CN111975768A (en) * | 2020-07-08 | 2020-11-24 | 华南理工大学 | Mechanical arm motion planning method based on fixed parameter neural network |
CN112894819A (en) * | 2021-01-29 | 2021-06-04 | 佛山树客智能机器人科技有限公司 | Robot dynamic motion control method and device based on double neural networks |
-
2021
- 2021-11-26 CN CN202111421980.9A patent/CN114167722B/en active Active
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2018176854A1 (en) * | 2017-03-27 | 2018-10-04 | 华南理工大学 | Method for programming repeating motion of redundant robotic arm |
CN110134013A (en) * | 2019-05-08 | 2019-08-16 | 杭州电子科技大学 | The parallel mechanical arm finite time convergence control motion control method of external disturbance can be fought |
CN111872933A (en) * | 2019-11-25 | 2020-11-03 | 浙江大学宁波理工学院 | SCARA robot trajectory tracking control method based on improved quadratic iterative learning control |
CN111443712A (en) * | 2020-03-30 | 2020-07-24 | 杭州电子科技大学 | Three-dimensional path planning method based on longicorn group search algorithm |
CN111975768A (en) * | 2020-07-08 | 2020-11-24 | 华南理工大学 | Mechanical arm motion planning method based on fixed parameter neural network |
CN112894819A (en) * | 2021-01-29 | 2021-06-04 | 佛山树客智能机器人科技有限公司 | Robot dynamic motion control method and device based on double neural networks |
Non-Patent Citations (1)
Title |
---|
彭志文;高宏力;梁超;文刚;: "RBF神经网络补偿的并联机器人控制研究", 机械设计与制造, no. 03, 8 March 2018 (2018-03-08) * |
Also Published As
Publication number | Publication date |
---|---|
CN114167722B (en) | 2024-03-29 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107490965B (en) | Multi-constraint trajectory planning method for space free floating mechanical arm | |
CN106945020B (en) | A kind of space double mechanical arms system motion control method for coordinating | |
Jung et al. | Force tracking impedance control of robot manipulators under unknown environment | |
CN112959325B (en) | High-precision control method for collaborative machining of double-moving mechanical arm in large scene | |
Wen et al. | Elman fuzzy adaptive control for obstacle avoidance of mobile robots using hybrid force/position incorporation | |
CN107962566B (en) | Repetitive motion planning method for mobile mechanical arm | |
US11498213B2 (en) | Robot joint space graph path planning and move execution | |
Tan et al. | Minimum time trajectory planner for the discrete dynamic robot model with dynamic constraints | |
CN111168684B (en) | On-orbit assembly sequence planning method for large-scale spatial structure | |
CN105598968B (en) | A kind of motion planning and control method of parallel mechanical arm | |
CN110076783A (en) | A kind of plane activation lacking mechanical arm position control method based on fuzzy control and liapunov function | |
CN108393886B (en) | Distributed multi-mobile manipulator cooperative carrying method capable of optimizing energy and operation degree | |
CN112091829B (en) | Sand blasting and rust removing parallel robot friction force mutation compensating fuzzy self-adaptive sliding mode control method | |
Dehio et al. | Modeling and control of multi-arm and multi-leg robots: Compensating for object dynamics during grasping | |
CN110598285A (en) | Method and device for solving inverse kinematics of manipulator track and storage medium | |
CN108908347A (en) | One kind is towards redundancy mobile mechanical arm error-tolerance type repetitive motion planning method | |
CN101143443A (en) | Six-freedom parallel robot decoupling method | |
CN109940615B (en) | Terminal state network optimization method for synchronous repeated motion planning of double-arm manipulator | |
CN109623812B (en) | Mechanical arm trajectory planning method considering spacecraft body attitude motion | |
CN114942593A (en) | Mechanical arm self-adaptive sliding mode control method based on disturbance observer compensation | |
CN114571444A (en) | Q-learning-based impedance control method for slagging-off robot | |
Kang et al. | Kinematic path‐tracking of mobile robot using iterative learning control | |
CN114167722A (en) | Parallel robot tracking control method based on super-exponential convergence neural network | |
JP2020049554A (en) | Track formation method, track formation device, and robot system | |
CN113103262A (en) | Robot control device and method for controlling robot |
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 |