CN106950823A - A kind of Neural network PID System with Sliding Mode Controller for intelligent wheel chair - Google Patents
A kind of Neural network PID System with Sliding Mode Controller for intelligent wheel chair Download PDFInfo
- Publication number
- CN106950823A CN106950823A CN201610005249.0A CN201610005249A CN106950823A CN 106950823 A CN106950823 A CN 106950823A CN 201610005249 A CN201610005249 A CN 201610005249A CN 106950823 A CN106950823 A CN 106950823A
- Authority
- CN
- China
- Prior art keywords
- neural network
- control
- sliding
- network pid
- wheel chair
- 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
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
- G05B13/041—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 variable is automatically adjusted to optimise the performance
Abstract
The invention discloses a kind of Neural network PID System with Sliding Mode Controller for intelligent wheel chair, it is related to the control field of intelligent wheel chair, it is characterised in that:Control structure includes master controller, motor drive module, the brshless DC motor with Hall element, driving wheel and wheelchair frame, master controller controls the rotating speed of direct current generator by Neural network PID sliding-mode control, the real-time speed feedback of motor to main controller, is formed closed loop speed control system by the Hall element on direct current generator.The present invention combines sliding formwork control and ANN Control, the PID sliding mode controllers based on neutral net for DC. machine design.Using the unknown load of Neural Network Science learning system and external disturbance, and by self adaptation moreNew powerValue makes system reach sliding-mode surface.Therefore, controller eliminated on the premise of robustness is not sacrificed and buffeted present in sliding formwork control, and the addition of PID sliding formworks can improve the stable state accuracy and dynamic property of intelligent wheel chair.
Description
Technical field
The present invention relates to the Neural network PID sliding formwork control side of the control field of intelligent wheel chair, more particularly to intelligent wheel chair
Method.
Background technology
China has become the most country of elderly population in the world, is also aging population country with the fastest developing speed
One of.Wheelchair is played an important role as the elderly and the basic walking-replacing tool of disadvantaged group in their daily life.
Manual wheelchair has the trend replaced by electric wheelchair due to some defects of itself.Electric wheelchair carries drive device, user
Only need manipulation bar, so that it may control the rotation and movement of wheelchair, but electric wheelchair is opened loop control, has had a strong impact on wheelchair
Running precision and stability.In this case, intelligent wheel chair arises at the historic moment, and not only function is more complete for intelligent wheel chair, operation speed
Degree is more accurate, and security is higher.
For the speed control of intelligent wheel chair, conventional control method is PID control method, though the method control is simple,
Stability is preferable, but its trackability is poor, particularly when the operation ground of wheelchair is not smooth enough, and its control accuracy is not often up to
To requiring, so for intelligent wheel chair, seek a kind of to take into account operation stability and the control method of control accuracy is particularly important.
The content of the invention
The purpose of the present invention is to overcome the deficiencies in the prior art to be slided there is provided a kind of Neural network PID for intelligent wheel chair
Mould control system, the addition of PID sliding formworks can improve the stable state accuracy of intelligent wheel chair, make intelligent wheel chair have good dynamic and
Static quality.
In order to solve the above-mentioned technical problem, the invention provides a kind of Neural network PID sliding formwork control for intelligent wheel chair
System processed, it is characterised in that:Control structure include master controller, motor drive module, the brshless DC motor with Hall element,
Driving wheel and wheelchair frame, direct current generator are solidified as a whole with driving wheel, and master controller sends control signal to motor drive module,
Hall element on the rotating speed of direct current generator, direct current generator is controlled by Neural network PID sliding-mode control by motor
Real-time speed feedback forms closed loop speed control system to main controller.
A kind of Neural network PID System with Sliding Mode Controller for intelligent wheel chair that the present invention is provided, its Neural network PID is slided
Mould control method includes step:
1)Set up the mechanical equation of direct current generator;
2)Neural network PID System with Sliding Mode Controller is set up, based on Neural network PID sliding formwork control design control law, as
The control input of direct current generator.Comprise the following steps:
2-1)Design PID sliding-mode surfacesFor:;
Wherein,It is the tracking error of direct current generator position output.It is DC generator speed
Tracking error.,For sliding formwork coefficient, and they are normal number;
2-2)Design Neural network PID sliding formwork control ratio, direct current generator actual path is tracked coideal track;
Wherein, the controller is:
In formula,It is mechanical index;It is damped coefficient,,,It is the rotation position of the rotor, the rotation of rotor respectively
Speed, the rotary acceleration of rotor,,Only be it is measurable,It is rotary inertia,It is normal number;
3)Using lyapunov function theories, adaptive law is designed, the progressive of the Neural network PID System with Sliding Mode Controller is verified
Stability.
Further, the step 1)In, the mechanical equation of brshless DC motor is:
In formula,It is mechanical index;It is damped coefficient,,,It is the rotation position of the rotor, the rotation speed of rotor
Degree, the rotary acceleration of rotor, respectively,,Only be it is measurable,It is rotary inertia;It is load torque and interference torque
Summation.
Further, the step 2)Middle neutral net is used for estimating load torque and the summation of disturbance torque,If
It is calculated as:,For mapping error.
Further, the lyapunov functions are designed as:, the adaptive rate is designed as: 。
Wherein,It is learning rate,It is the weight vectors in neutral net,It is Gaussian function,It is estimative weight vectors,It is estimative weight vectors
Error,。
Implement the present invention, have the advantages that:1st, a kind of Neural network PID sliding formwork control system for intelligent wheel chair
System, because sliding mode can be designed and parameter with direct current generator and disturbance are unrelated, this, which allows for sliding formwork control, has
Quick response, to Parameters variation and insensitive, the advantages of physics realization is simple of disturbance.This control method is cut by controlled quentity controlled variable
Changing makes system mode be slided along sliding-mode surface, makes the control system of intelligent wheel chair when by Parameter Perturbation and external interference
With consistency.
2nd, the addition of PID sliding formworks can improve the stable state accuracy of intelligent wheelchair control system, have intelligent wheel chair good
Dynamic and static performance.
3rd, adaptive law is designed based on Lyapunov stability theory, realizes the online updating of parameter, can be in office
In the case of meaning initial value, it is ensured that the Global asymptotic stability of intelligent wheelchair control system.
Brief description of the drawings
Fig. 1 is intelligent wheelchair control system functional-block diagram of the invention.
Fig. 2 is brshless DC motor control block diagram in the present invention.
Embodiment
For the ease of the understanding of the present invention, this is described further.
The invention provides a kind of Neural network PID System with Sliding Mode Controller for intelligent wheel chair, Fig. 1 is present invention intelligence
Wheelchair control system functional-block diagram, it is characterised in that:Control structure includes master controller, motor drive module, band Hall member
Brshless DC motor, driving wheel and the wheelchair frame of part, direct current generator are solidified as a whole with driving wheel, and master controller sends control letter
Number give motor drive module, the rotating speed of direct current generator is controlled by Neural network PID sliding-mode control, meanwhile, direct current
The real-time speed feedback of motor to main controller, is formed closed loop speed control system by the Hall element on machine.
Fig. 2 is the brshless DC motor control block diagram of intelligent wheel chair of the present invention, using Neural network PID sliding formwork control side
Method, comprises the following steps:
1. set up the mechanical equation of brshless DC motor:
. (1)
In formula,It is mechanical index;It is damped coefficient,,,It is the rotation position of the rotor, the rotation speed of rotor
Degree, the rotary acceleration of rotor, respectively,,Only be it is measurable,It is rotary inertia;It is load torque and interference torque
Summation;
2. set up Neural network PID System with Sliding Mode Controller, based on Neural network PID sliding formwork control design control law, as
The control input of direct current generator.
Define tracking errorFor:. (2)
Wherein,For position command,For the actual rotation position of DC motor rotor.
Design PID sliding-mode surfaces, make sliding mode asymptotically stability determined by it and with good dynamic quality.S is designed
For: (3)
Wherein,It is the tracking error of direct current generator position output,It is the tracking mistake of speed
Difference,,For sliding formwork coefficient, and they are normal number.
Design Neural network PID sliding formwork control ratio, direct current generator actual path is tracked coideal track.
Wherein, the controller is:
. (4)
In formula,It is mechanical index;It is damped coefficient,,,It is the rotation position of the rotor, the rotation speed of rotor
Degree, the rotary acceleration of rotor,,Only be it is measurable,It is rotary inertia,It is normal number.
3. using lyapunov function theories, adaptive law is designed, the Neural network PID System with Sliding Mode Controller is verified
Asymptotic Stability.
Lyapunov functions are designed as:
. (5)
Wherein,It is learning rate,It is the weight vectors in neutral net,It is Gaussian function.It is estimative weight vectors,Estimative weight to
The error of amount,。
To PID sliding-mode surfacesDerivation, is obtained:
. (6)
Neutral net is used for estimating load torque and the summation of disturbance torque,It is designed as:
. (7)
Wherein,For the mapping error of neutral net.
By Neural network PID sliding formwork control ratio formula(4)Formula is brought into as the control input of direct current generator(6), obtain
. (8)
So
. (9)
Adaptive rate is designed as:
. (10)
Then have:
. (11)
Because,All it is positive number,It is negative semidefinite, so the control system of the intelligent wheel chair is stable.
A kind of Neural network PID System with Sliding Mode Controller for intelligent wheel chair that the present invention is provided, because sliding mode can
To be designed and parameter with direct current generator and disturbance are unrelated, this, which allows for sliding formwork control, has quick response, and parameter is become
Change and disturb insensitive, the advantages of physics realization is simple.This control method by the switching of controlled quentity controlled variable make system mode along
Sliding-mode surface is slided, and make the control system of intelligent wheel chair has consistency when by Parameter Perturbation and external interference.PID is slided
The addition of mould can improve the stable state accuracy of intelligent wheelchair control system, intelligent wheel chair is had good dynamic and static product
Matter.Adaptive law is designed based on Lyapunov stability theory, the online updating of parameter is realized, can be in any initial value
In the case of, it is ensured that the Global asymptotic stability of intelligent wheelchair control system.
Claims (4)
1. a kind of Neural network PID System with Sliding Mode Controller for intelligent wheel chair, it is characterised in that:Control structure includes main control
Device, motor drive module, the brshless DC motor with Hall element, driving wheel and wheelchair frame, direct current generator are consolidated with driving wheel
It is integrated, master controller sends control signal to motor drive module, is controlled by Neural network PID sliding-mode control straight
The rotating speed of motor is flowed, meanwhile, the real-time speed feedback of motor to main controller, is formed speed and closed by the Hall element on direct current generator
Ring control system.
2. a kind of Neural network PID System with Sliding Mode Controller for intelligent wheel chair according to claim 1, its feature exists
In:Neural network PID sliding-mode control, comprises the following steps:
1)Set up the mechanical equation of direct current generator;
2)Neural network PID System with Sliding Mode Controller is set up, based on Neural network PID sliding formwork control design control law, as
The control input of direct current generator, comprises the following steps:
2-1)Design PID sliding-mode surfacesFor:
Wherein,It is the tracking error of direct current generator outgoing position,Be motor speed with
Track error,,For sliding formwork coefficient, and they are normal number;
2-2)Design Neural network PID sliding formwork control ratio, direct current generator actual path is tracked coideal track;
Wherein, the controller is:
Wherein,It is mechanical index;It is damped coefficient,,,It is the rotation position of the rotor, the rotary speed of rotor,
The rotary acceleration of rotor, respectively,,Only be it is measurable,It is rotary inertia,It is normal number;
3)Using lyapunov function theories, adaptive law is designed, the progressive of the Neural network PID System with Sliding Mode Controller is verified
Stability;
The lyapunov functions are designed as:
Adaptive rate is designed as:
Wherein,It is learning rate,It is the weight vectors in neutral net,It is estimative weight vectorsIt is the error of estimative weight vectors,It is Gaussian function,。
3. a kind of Neural network PID System with Sliding Mode Controller for intelligent wheel chair according to claim 2, its feature exists
In the step 1)In, the mechanical equation of brshless DC motor is:
In formula,It is mechanical index;It is damped coefficient,,,It is the rotation position of the rotor, the rotary speed of rotor,
The rotary acceleration of rotor, respectively,,Only be it is measurable,It is rotary inertia;It is load torque and disturbs the total of torque
With.
4. a kind of Neural network PID System with Sliding Mode Controller for intelligent wheel chair according to claim 2, it is characterised in that
Neutral net is used for estimating load torque and the summation of disturbance torque, i.e.,;
The summation of load torque and disturbance torqueIt is designed as:
,For mapping error.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610005249.0A CN106950823A (en) | 2016-01-07 | 2016-01-07 | A kind of Neural network PID System with Sliding Mode Controller for intelligent wheel chair |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201610005249.0A CN106950823A (en) | 2016-01-07 | 2016-01-07 | A kind of Neural network PID System with Sliding Mode Controller for intelligent wheel chair |
Publications (1)
Publication Number | Publication Date |
---|---|
CN106950823A true CN106950823A (en) | 2017-07-14 |
Family
ID=59466005
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201610005249.0A Pending CN106950823A (en) | 2016-01-07 | 2016-01-07 | A kind of Neural network PID System with Sliding Mode Controller for intelligent wheel chair |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN106950823A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107544254A (en) * | 2017-10-12 | 2018-01-05 | 北京航空航天大学 | A kind of adaptive dynamic sliding mode control method for it is expected margin of safety following-speed model |
CN108227491A (en) * | 2017-12-28 | 2018-06-29 | 重庆邮电大学 | A kind of intelligent vehicle Trajectory Tracking Control method based on sliding formwork neural network |
CN108227770A (en) * | 2018-01-03 | 2018-06-29 | 深圳市易成自动驾驶技术有限公司北京分公司 | Wheelchair drives computational methods, device and the computer readable storage medium of rotating speed |
CN108566086A (en) * | 2018-04-13 | 2018-09-21 | 杭州电子科技大学 | Two close cycles RBF neural sliding moding structure adaptive control system |
-
2016
- 2016-01-07 CN CN201610005249.0A patent/CN106950823A/en active Pending
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107544254A (en) * | 2017-10-12 | 2018-01-05 | 北京航空航天大学 | A kind of adaptive dynamic sliding mode control method for it is expected margin of safety following-speed model |
CN107544254B (en) * | 2017-10-12 | 2020-04-14 | 北京航空航天大学 | Adaptive dynamic sliding mode control method of expected safety margin following model |
CN108227491A (en) * | 2017-12-28 | 2018-06-29 | 重庆邮电大学 | A kind of intelligent vehicle Trajectory Tracking Control method based on sliding formwork neural network |
CN108227770A (en) * | 2018-01-03 | 2018-06-29 | 深圳市易成自动驾驶技术有限公司北京分公司 | Wheelchair drives computational methods, device and the computer readable storage medium of rotating speed |
CN108566086A (en) * | 2018-04-13 | 2018-09-21 | 杭州电子科技大学 | Two close cycles RBF neural sliding moding structure adaptive control system |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102385342B (en) | Self-adaptation dynamic sliding mode controlling method controlled by virtual axis lathe parallel connection mechanism motion | |
CN108233781B (en) | Direct current motor self-adaptive inversion sliding mode control method based on disturbance observer | |
CN106950823A (en) | A kind of Neural network PID System with Sliding Mode Controller for intelligent wheel chair | |
CN102621889B (en) | Composite control method for piezoelectric ceramic positioning | |
GB2572458A (en) | An adaptive sliding-mode control method of the permanent magnet synchronous motor based on dynamic surface technology | |
CN109927032A (en) | A kind of mechanical arm Trajectory Tracking Control method based on High-Order Sliding Mode observer | |
CN104300863B (en) | A kind of adaptive sliding-mode observer method of varying load PMSM Speed | |
CN106788044A (en) | A kind of permagnetic synchronous motor self adaptation non-singular terminal sliding-mode control based on interference observer | |
CN107207040B (en) | Electric power steering apparatus | |
CN106647283A (en) | Auto-disturbance rejection position servo system optimization design method based on improved CPSO | |
Tatikonda et al. | Control of inverted pendulum using adaptive neuro fuzzy inference structure (ANFIS) | |
CN109067264A (en) | A kind of balance car system and its control method | |
CN103034126A (en) | Controlling system and controlling method of axial off-center magnetic bearing of outer rotor of constant current source | |
CN107248822A (en) | Inverter control method based on Fractional Order PID discrete sliding mode structure changes | |
Kiruthika et al. | Mathematical modelling and speed control of a sensored brushless DC motor using intelligent controller | |
CN106707753A (en) | Pump linear motor adaptive control method | |
CN107045285A (en) | A kind of servo system self-adaptive parameter identification and control method with input saturation | |
CN106549609A (en) | Method for controlling permanent magnet synchronous motor based on integral form high order termination sliding Mode Algorithm | |
CN106877769B (en) | A kind of method of servo motor plus of speed controller parameter self-tuning | |
CN107528514A (en) | The Approximation Discrete fast terminal sliding-mode control of PMSM governing systems | |
Shi et al. | The research of fuzzy PID control application in DC motor of automatic doors | |
Zhao et al. | Proportional-integral based fuzzy sliding mode control of the milling head | |
CN107017820A (en) | A kind of switched reluctance machines vibration and noise reducing system | |
CN110649845A (en) | Photoelectric turntable position tracking control method based on robust generalized predictive control | |
CN104022701B (en) | Mould method for control speed in a kind of permanent magnetic linear synchronous motor Newton method |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170714 |
|
WD01 | Invention patent application deemed withdrawn after publication |