CN114460845A - Delta manipulator control method added with CMAC uncertainty compensation - Google Patents

Delta manipulator control method added with CMAC uncertainty compensation Download PDF

Info

Publication number
CN114460845A
CN114460845A CN202210083189.XA CN202210083189A CN114460845A CN 114460845 A CN114460845 A CN 114460845A CN 202210083189 A CN202210083189 A CN 202210083189A CN 114460845 A CN114460845 A CN 114460845A
Authority
CN
China
Prior art keywords
cmac
compensation
uncertainty
neural network
manipulator
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
Application number
CN202210083189.XA
Other languages
Chinese (zh)
Inventor
彭志文
朱鹏
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
724th Research Institute of CSIC
Original Assignee
724th Research Institute of CSIC
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 724th Research Institute of CSIC filed Critical 724th Research Institute of CSIC
Priority to CN202210083189.XA priority Critical patent/CN114460845A/en
Publication of CN114460845A publication Critical patent/CN114460845A/en
Pending legal-status Critical Current

Links

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
    • 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

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Evolutionary Computation (AREA)
  • Health & Medical Sciences (AREA)
  • Theoretical Computer Science (AREA)
  • Artificial Intelligence (AREA)
  • Software Systems (AREA)
  • General Physics & Mathematics (AREA)
  • General Health & Medical Sciences (AREA)
  • Mathematical Physics (AREA)
  • Computational Linguistics (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Biophysics (AREA)
  • Data Mining & Analysis (AREA)
  • Biomedical Technology (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Medical Informatics (AREA)
  • Automation & Control Theory (AREA)
  • Feedback Control In General (AREA)

Abstract

The invention relates to a Delta manipulator control method added with CMAC neural network uncertainty compensation, belonging to the field of manipulator control. The Delta manipulator control is based on a manipulator dynamics model, a CMAC neural network is designed to approximate and compensate uncertainty on the basis of analyzing the uncertainty of the model, the input of the neural network is the position and speed errors of three main arms of the manipulator, the output is the moment compensation quantity of the three main arms, and the output quantity of a calculated moment control algorithm is superposed with the compensation quantity of the CMAC neural network to be jointly used as the control output of the Delta manipulator. The method designed by the invention can effectively improve the motion precision of the manipulator while ensuring the stability of the control system, and provides an effective control method for the field of manipulator control.

Description

Delta manipulator control method added with CMAC uncertainty compensation
Technical Field
The invention relates to the field of manipulator control.
Background
In recent years, parallel manipulators represented by Delta manipulators have the advantages of higher rigidity, no accumulated errors of joint mechanical arms, higher accuracy and the like, and are widely applied to industries such as medical treatment, automatic assembly production lines, defective product sorting, 3D printing and the like. The control stability and the control precision of the Delta parallel manipulator directly influence the working efficiency and the application range of a production line, and the research on the Delta parallel manipulator control method has obvious practical significance. Compared with a serial mechanical arm which is more mature in application, the Delta parallel mechanical arm has high nonlinearity and coupling performance, and the linear control strategies such as a PD controller and the like which are widely applied at present are adopted, the linear controller treats all branched chains of a system as independent branches to carry out independent control, although the calculation amount is small and the control is simple, the control precision is insufficient, and the complicated application working condition cannot be met.
At present, the moment control based on the mechanical arm dynamic model is widely applied. However, since the dynamic model of the manipulator is obtained in an ideal state, uncertain factors such as load variation, inertia fluctuation, joint friction and the like exist in the actual working condition, and an ideal calculation torque control strategy is difficult to be applied to complex working conditions, especially variable load working conditions.
Disclosure of Invention
In order to solve the problems that accurate modeling is difficult and load time variation is difficult in a parallel manipulator dynamics control method, the invention provides a Delta manipulator control method with CMAC uncertainty compensation, which can estimate uncertainty in the manipulator control process and can be added into a controller to improve the stability and robustness of manipulator control.
The invention is realized by adopting the following technical scheme:
the Delta manipulator control method added with CMAC uncertainty compensation comprises the following steps:
s1: establishing an inverse kinematics model and an ideal dynamics model of the Delta manipulator system;
s2: analyzing uncertainty analysis of the Delta manipulator system;
s3: designing a CMAC neural network for approximating the uncertainty of the model;
s4: and adding a neural network compensation algorithm based on a calculation moment control strategy to obtain a control strategy added with CMAC uncertainty compensation.
Further, the step S4 of adding the CMAC uncertainty compensation control strategy includes: the neural network input layer is composed of angle errors of three driving joints and derivatives thereof, and comprises 6 neurons; the hidden layer comprises 5 nodes; the output layer is a torque compensation value of 3 driving arms of the Delta manipulator; the neural network is compensated on line, the neural network is trained in the real-time tracking process of the tail end track of the robot, the weight of the network is corrected on line according to the track position and the speed error, the optimal output is continuously approached, and the off-line training process is avoided; the control law of adding CMAC network online compensation is as follows:
Figure BDA0003486720810000021
wherein: t is the drive moment of the drive arm; m (α) is a mass inertia matrix;
Figure BDA0003486720810000022
is a matrix of coriolis and centripetal forces; n (α) is a gravity matrix; alpha is the vector of the rotation angle of the driving arm; f. ofeIs the CMAC network on-line compensation torque value.
By adopting the technical approach, the invention has the following beneficial effects:
the Delta manipulator control method added with the CMAC neural network compensation analyzes the uncertainty of a system dynamic model, designs the CMAC neural network on-line compensation based on a simple calculation torque control strategy, and forms an algorithm which is compared with the calculation torque control strategy which depends on the accuracy of the model seriously, so that the method can adapt to more complicated and changeable working conditions, and has higher control precision, lower control error and better effect.
Drawings
FIG. 1 is a design flow chart of a Delta manipulator control method;
FIG. 2 is a block diagram of a CMAC neural network compensation control algorithm.
Detailed Description
The invention is further explained below with reference to the drawings.
The invention provides a Delta manipulator control method with CMAC uncertainty compensation, which comprises the following steps:
and S1, establishing an inverse kinematics model and an ideal dynamics model of the Delta manipulator system.
The ideal dynamic model of the Delta manipulator system is as follows:
Figure BDA0003486720810000023
wherein:
m (α): quality of foodA magnitude inertia matrix;
Figure BDA0003486720810000024
a kirschner force and centripetal force matrix; n (α): a gravity matrix; α: a drive arm rotation angle vector; t: a drive arm drive torque; t isd: and (4) external disturbance.
And S2, analyzing the uncertainty of the system.
Firstly, the dynamic model of the system is assumed to be accurately established, and the ideal model of the system is consistent with the real model and has no external disturbance. Then the following relationship exists:
Figure BDA0003486720810000025
the moment control strategy based on dynamics design is as follows:
Figure BDA0003486720810000026
the closed-loop error equation of the system can be obtained by substituting the formula (2) into the formula (1)
Figure BDA0003486720810000027
In actual conditions, a complete and real dynamic model cannot be obtained, and various determined and uncertain external disturbances exist. Then the following deviations exist between the ideal model and the system reality model:
Figure BDA0003486720810000028
therefore, in a real situation, the closed-loop error equation of the system is:
Figure BDA0003486720810000029
i.e. the uncertainty of the system.
And S3, designing the CMAC neural network to approximate the uncertainty of the model.
Aiming at the step 3, the designed CMAC neural network selects a Gaussian function as a basis function, and the input of the CMAC neural network is the angle errors and derivatives of the three driving joints; the CMAC network output is the torque compensation value for the three drive arms.
And S4, adding a CMAC neural network compensation algorithm based on the calculation torque control strategy to obtain a control strategy added with CMAC uncertainty compensation.
Aiming at the step 4, the control rate for adding the compensation of the CMAC neural network is designed as follows:
Figure BDA0003486720810000031
order to
Figure BDA0003486720810000032
Then T is equal to T1+T2。
Wherein T is1Indicating calculated torque control quantity, T2And expressing the uncertainty compensation control quantity of the CMAC neural network.
Figure BDA0003486720810000033
The self-adaptive law of the network weight is taken as follows:
Figure BDA0003486720810000034
in the formula:
Figure BDA0003486720810000035
in order to be able to output the desired output,
Figure BDA0003486720810000036
α is a learning constant for actual output.

Claims (2)

1. The Delta manipulator control method added with CMAC uncertainty compensation is characterized in that:
s1: establishing an inverse kinematics model and an ideal dynamics model of the Delta manipulator system;
s2: analyzing system uncertainty due to load variations, inertia fluctuations, and dimensional non-uniformities;
s3: designing a CMAC neural network for approximating the uncertainty of the model;
s4: and adding a neural network compensation algorithm based on a calculation moment control strategy to obtain a control strategy added with CMAC uncertainty compensation.
2. The method of Delta manipulator control incorporating CMAC uncertainty compensation of claim 1, wherein: the control strategy of adding CMAC uncertainty compensation in S4 comprises: the neural network input layer is composed of angle errors of three driving joints and derivatives thereof, and comprises 6 neurons; the hidden layer comprises 5 nodes; the output layer is a torque compensation value of 3 driving arms of the Delta manipulator; the neural network is compensated on line, the neural network is trained in the real-time tracking process of the tail end track of the robot, the weight of the network is corrected on line according to the track position and the speed error, the optimal output is continuously approached, and the off-line training process is avoided;
the control law of adding CMAC network online compensation is as follows:
Figure FDA0003486720800000011
wherein: t is the drive moment of the drive arm; m (α) is a mass inertia matrix;
Figure FDA0003486720800000012
is a matrix of coriolis and centripetal forces; n (α) is a gravity matrix; alpha is the vector of the rotation angle of the driving arm; f. ofeIs the CMAC network on-line compensation torque value.
CN202210083189.XA 2022-01-25 2022-01-25 Delta manipulator control method added with CMAC uncertainty compensation Pending CN114460845A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202210083189.XA CN114460845A (en) 2022-01-25 2022-01-25 Delta manipulator control method added with CMAC uncertainty compensation

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202210083189.XA CN114460845A (en) 2022-01-25 2022-01-25 Delta manipulator control method added with CMAC uncertainty compensation

Publications (1)

Publication Number Publication Date
CN114460845A true CN114460845A (en) 2022-05-10

Family

ID=81412251

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202210083189.XA Pending CN114460845A (en) 2022-01-25 2022-01-25 Delta manipulator control method added with CMAC uncertainty compensation

Country Status (1)

Country Link
CN (1) CN114460845A (en)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB9216952D0 (en) * 1991-08-14 1992-09-23 Toshiba Kk Predictive control method and apparatus
CN108942924A (en) * 2018-06-25 2018-12-07 南京理工大学 Model uncertainty mechanical arm motion control method based on multilayer neural network
CN109176525A (en) * 2018-09-30 2019-01-11 上海神添实业有限公司 A kind of mobile manipulator self-adaptation control method based on RBF
CN111185907A (en) * 2020-01-13 2020-05-22 福州大学 Pose stability control method for operation type flying robot after grabbing
CN111872937A (en) * 2020-07-23 2020-11-03 西华大学 Control method for uncertain mechanical arm in task space

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
GB9216952D0 (en) * 1991-08-14 1992-09-23 Toshiba Kk Predictive control method and apparatus
CN108942924A (en) * 2018-06-25 2018-12-07 南京理工大学 Model uncertainty mechanical arm motion control method based on multilayer neural network
CN109176525A (en) * 2018-09-30 2019-01-11 上海神添实业有限公司 A kind of mobile manipulator self-adaptation control method based on RBF
CN111185907A (en) * 2020-01-13 2020-05-22 福州大学 Pose stability control method for operation type flying robot after grabbing
CN111872937A (en) * 2020-07-23 2020-11-03 西华大学 Control method for uncertain mechanical arm in task space

Non-Patent Citations (2)

* Cited by examiner, † Cited by third party
Title
于乃功: "基于CMAC 神经网络的机械手臂实时控制", 《中南大学学报(自然科学版)》, vol. 38, pages 526 - 529 *
吕栋腾: "基于神经网络的火电厂脱硫控制系统研究", 《机械与电子》, vol. 39, no. 9, pages 37 - 40 *

Similar Documents

Publication Publication Date Title
CN109676607B (en) Zero gravity control method without torque sensing
CN110238839B (en) Multi-shaft-hole assembly control method for optimizing non-model robot by utilizing environment prediction
CN105773623B (en) SCARA robotic tracking control methods based on the study of forecasting type Indirect iteration
CN105772917B (en) A kind of three joint spot welding robot's Trajectory Tracking Control methods
CN109782601B (en) Design method of self-adaptive neural network synchronous robust controller of coordinated mechanical arm
CN108942924A (en) Model uncertainty mechanical arm motion control method based on multilayer neural network
CN109176525A (en) A kind of mobile manipulator self-adaptation control method based on RBF
CN112518753B (en) Industrial robot trajectory tracking system and method based on neural network iterative compensation
CN111872937B (en) Control method for uncertain mechanical arm in task space
US20220371186A1 (en) Dual-robot position/force multivariate-data-driven method using reinforcement learning
CN113427483A (en) Double-machine manpower/bit multivariate data driving method based on reinforcement learning
CN111687827A (en) Control method and control system for coordinating and operating weak rigid member by two robots
CN110861090B (en) Torque feedforward control system and method
CN108227506A (en) A kind of robot admittance control system based on adaptive optimization method
CN109648564B (en) Control method of multi-degree-of-freedom flexible joint mechanical arm system based on hierarchical MPC (Multi-degree-of-freedom)
CN111428317A (en) Joint friction torque compensation method based on 5G and recurrent neural network
CN114536334A (en) High-order sliding mode anti-interference control method for flexible mechanical arm system
CN114347018A (en) Mechanical arm disturbance compensation method based on wavelet neural network
CN114310851A (en) Robot dragging-free teaching method for torque sensor
CN106607901A (en) Joint controller of three-degree-of-freedom pneumatic series connection mechanical arm
Gao et al. Time-optimal trajectory planning of industrial robots based on particle swarm optimization
CN109176480A (en) A kind of sliding-mode control and system of parallel robot
WO2023020036A1 (en) Redundant manipulator tracking control method based on echo state network
CN113325716B (en) Underwater hydraulic mechanical arm nonlinear robust control method based on extended observer
CN114460845A (en) Delta manipulator control method added with CMAC uncertainty compensation

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
CB02 Change of applicant information

Address after: 210003 No. 346, Zhongshan North Road, Jiangsu, Nanjing

Applicant after: 724 Research Institute of China Shipbuilding Corp.

Address before: 210003 No. 346, Zhongshan North Road, Jiangsu, Nanjing

Applicant before: 724TH RESEARCH INSTITUTE OF CHINA SHIPBUILDING INDUSTRY Corp.

CB02 Change of applicant information
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20220510

WD01 Invention patent application deemed withdrawn after publication