CN106877746B - Method for control speed and speed control unit - Google Patents
Method for control speed and speed control unit Download PDFInfo
- Publication number
- CN106877746B CN106877746B CN201710167446.7A CN201710167446A CN106877746B CN 106877746 B CN106877746 B CN 106877746B CN 201710167446 A CN201710167446 A CN 201710167446A CN 106877746 B CN106877746 B CN 106877746B
- Authority
- CN
- China
- Prior art keywords
- speed
- neural network
- training data
- network model
- instruction
- 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
Classifications
-
- H—ELECTRICITY
- H02—GENERATION; CONVERSION OR DISTRIBUTION OF ELECTRIC POWER
- H02P—CONTROL OR REGULATION OF ELECTRIC MOTORS, ELECTRIC GENERATORS OR DYNAMO-ELECTRIC CONVERTERS; CONTROLLING TRANSFORMERS, REACTORS OR CHOKE COILS
- H02P5/00—Arrangements specially adapted for regulating or controlling the speed or torque of two or more electric motors
Landscapes
- Engineering & Computer Science (AREA)
- Power Engineering (AREA)
- Feedback Control In General (AREA)
Abstract
The invention discloses a kind of method for control speed and speed control units, are related to field of computer technology.Nonlinear neural network model in the present invention can simulate the non-linear relation of speed command and velocity amplitude under varying environment, by the way that modified speed command can be obtained with historical speed value unbalanced input neural network model in the historical speed instruction in present speed instruction and preset time, modified speed command is to be modified according to the non-linear relation of speed command and velocity amplitude to present speed instruction, again desired velocity amplitude will can be then obtained in the modified system of the speed command input based on PID control, improve the accuracy and precision of speed control.
Description
Technical field
The present invention relates to field of computer technology, in particular to a kind of method for control speed and speed control unit.
Background technique
In the case where the Chinese government carries forward vigorously industrialization and information-based two change the driving greatly of fusion, material flow industry revolution is quiet
It so rises, major logistics company increases the investment of wisdom logistics one after another, and in artificial intelligence, robot, the fields such as machine learning are not
Disconnected absorption, combination, innovation.The transfer robot in unmanned storehouse promotes the emphasis equipment of access efficiency as logistics cost is reduced,
Method for control speed directly influences the stability and working efficiency of robot.
Existing unmanned storehouse transfer robot mostly uses greatly Dual-motors Driving scheme, and the motor of each driving wheel carries out independent
Speed planning and control, using traditional PID (Proportion Integration Differentiation, proportional integration
Differential) controller of the controller as speed closed loop.It is this using single PID controller transfer robot to be controlled
Servo drive system is reduced to linear time invariant system by scheme.But in fact, with the variation of environment, such as continuous operation
Time, reasons, the system performance such as load frequent variation are changing always, might not meet linear rule, and PID control
The control parameter of device cannot but adjust in real time, this is allowed for for the precision of the speed control of storehouse transfer robot and accurate
It spends lower.
Summary of the invention
A technical problem to be solved by this invention is: how to improve the essence of the speed control for storehouse transfer robot
Degree and accuracy.
According to one embodiment of present invention, a kind of method for control speed provided, comprising: by present speed instruction, in advance
If the historical speed instruction in the time and the historical speed in preset time instruct corresponding historical speed value unbalanced input
Neural network model；The erection rate instruction input ratio corresponding with present speed instruction that nonlinear neural network model is exported
Example integral differential PID controller, so that PID controller controls speed according to erection rate instruction.
In one embodiment, this method further include: acquire the historical speed instruction and different operating conditions under different operating conditions
Under historical speed instruct corresponding historical speed value as training data；Training data is normalized；Utilize normalization
Training data determine the parameter of nonlinear neural network model, so that it is determined that nonlinear neural network model.
In one embodiment, training data is normalized includes: by the training data before normalization and to train number
According to the difference of minimum value after the poor training data divided by after normalization of middle minimum value and default normalization as the first ratio；It will return
One change before training data in maxima and minima difference divided by the difference of maxima and minima after default normalization as the
Two ratios；Training data equal with the second ratio according to the first ratio, after determining normalization.
In one embodiment, this method further include: according to feeding back to non-linear neural in real time or every predetermined period
The speed command and velocity amplitude of network model, are modified nonlinear neural network model.
In one embodiment, PID controller is using preset scale parameter, integral parameter, differential parameter to amendment speed
The velocity amplitude for the last moment that degree instruction and feedback obtain carries out operation, rate controlling amount is determined, to control to speed
System.
In one embodiment, nonlinear neural network is non-linear active autoregression NARX neural network, NARX nerve
The hidden neuron of network is 8, and delay order is 2.
According to another embodiment of the invention, a kind of speed control unit provided, comprising: neural network module is used
In present speed is instructed, the historical speed instruction in the instruction of historical speed in preset time and preset time is corresponding goes through
History velocity amplitude unbalanced input neural network model；Pid control module, for exporting nonlinear neural network model and working as
The corresponding erection rate instruction input proportional integral differential PID controller of preceding speed command, so that PID controller is according to amendment speed
Degree instruction controls speed.
In one embodiment, neural network module is also used to acquire the instruction of the historical speed under different operating conditions and difference
Historical speed under operating condition instructs corresponding historical speed value as training data, and training data is normalized, using returning
One training data changed determines the parameter of nonlinear neural network model, so that it is determined that nonlinear neural network model.
In one embodiment, after by the difference of minimum value in the training data and training data before normalization divided by normalization
Training data and default normalization after minimum value difference as the first ratio, by maximum value in the training data before normalization with
The difference of minimum value divided by the difference of maxima and minima after default normalization as the second ratio, by making the first ratio and the
Training data after the equal determining normalization of two ratios.
In one embodiment, neural network module is also used to non-linear according to feeding back in real time or every predetermined period
The speed command and velocity amplitude of neural network model, are modified nonlinear neural network model.
In one embodiment, pid control module is also using preset scale parameter, integral parameter, differential parameter to repairing
The velocity amplitude for the last moment that positive speed command and feedback obtain carries out operation, determines rate controlling amount, so as to speed into
Row control.
In one embodiment, nonlinear neural network is non-linear active autoregression NARX neural network, NARX nerve
The hidden neuron of network is 8, and delay order is 2.
According to still another embodiment of the invention, a kind of speed control unit provided, comprising: memory；And coupling
To the processor of memory, processor is configured as based on the instruction being stored in memory devices, execute as it is aforementioned any one
Method for control speed in embodiment.
A kind of still another embodiment in accordance with the present invention, the computer readable storage medium provided, is stored thereon with calculating
Machine program, which is characterized in that the program realizes the method for control speed in any one aforementioned embodiment when being executed by processor
Step.
Nonlinear neural network model in the present invention can simulate the non-thread of speed command and velocity amplitude under varying environment
Sexual intercourse, by instructing and historical speed value unbalanced input mind the historical speed in present speed instruction and preset time
It can be obtained modified speed command through network model, modified speed command is to be according to the non-thread of speed command and velocity amplitude
Sexual intercourse is modified present speed instruction, then the modified speed command is inputted the system based on PID control
In can then obtain desired velocity amplitude, improve the accuracy and precision of speed control.
By referring to the drawings to the detailed description of exemplary embodiment of the present invention, other feature of the invention and its
Advantage will become apparent.
Detailed description of the invention
In order to more clearly explain the embodiment of the invention or the technical proposal in the existing technology, to embodiment or will show below
There is attached drawing needed in technical description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention for those of ordinary skill in the art without creative efforts, can be with
It obtains other drawings based on these drawings.
Fig. 1 shows the structural schematic diagram of the speed control unit of one embodiment of the present of invention.
Fig. 2 shows the structural schematic diagrams of the speed control unit of another embodiment of the invention.
Fig. 3 shows the flow diagram of the method for control speed of one embodiment of the present of invention.
Fig. 4 shows the transfer robot chassis wheel distribution schematic diagram of one embodiment of the present of invention.
Fig. 5 shows the flow diagram of the method for control speed of another embodiment of the invention.
Fig. 6 shows the flow diagram of the method for control speed of another embodiment of the invention.
Fig. 7 shows the flow diagram of the method for control speed of an application examples of the invention.
Fig. 8 shows the structural schematic diagram of the speed control unit of another embodiment of the invention.
Specific embodiment
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, complete
Site preparation description, it is clear that described embodiments are only a part of the embodiments of the present invention, instead of all the embodiments.Below
Description only actually at least one exemplary embodiment be it is illustrative, never as to the present invention and its application or make
Any restrictions.Based on the embodiments of the present invention, those of ordinary skill in the art are not making creative work premise
Under every other embodiment obtained, shall fall within the protection scope of the present invention.
For using in the prior art, PID controller carries out control accuracy to the speed of transfer robot and precision compares
Low problem proposes this programme.
Speed control unit in the embodiment of the present invention can realize respectively by various calculating equipment or computer system, under
Face combines Fig. 1 and Fig. 2 to be described.
Fig. 1 is the structure chart of one embodiment of speed control unit of the present invention.As shown in Figure 1, the device of the embodiment
10 include: memory 110 and the processor 120 for being coupled to the memory 110, and processor 120 is configured as being based on being stored in
Instruction in memory 110 executes the method for control speed in the present invention in any one embodiment.
Wherein, memory 110 is such as may include system storage, fixed non-volatile memory medium.System storage
Device is for example stored with operating system, application program, Boot loader (Boot Loader), database and other programs etc..
Fig. 2 is the structure chart of another embodiment of speed control unit of the present invention.As shown in Fig. 2, the dress of the embodiment
Setting 10 includes: memory 110 and processor 120, can also be connect including input/output interface 230, network interface 240, storage
Mouth 250 etc..It can for example be connected by bus 260 between these interfaces 230,240,250 and memory 110 and processor 120
It connects.Wherein, input/output interface 230 is display, the input-output equipment such as mouse, keyboard, touch screen provide connecting interface.Net
Network interface 240 provides connecting interface for various networked devices, such as may be coupled to database server or cloud storage clothes
Business device etc..The external storages such as memory interface 250 is SD card, USB flash disk provide connecting interface.
Method for control speed of the invention is described below with reference to Fig. 3.Method of the invention can be applied to automated warehousing
In for transfer robot speed control, also can be applied to the control of other automated mechanicals, speed is also not limited to row
Walking speed also may include other speed such as rotation speed.
Fig. 3 is the flow chart of method for control speed one embodiment of the present invention.As shown in figure 3, the method packet of the embodiment
It includes:
Step S302, by the historical speed instruction in present speed instruction, preset time and the history in preset time
The corresponding historical speed value unbalanced input neural network model of speed command.
Include desired velocity amplitude in speed command, can be issued by Central Control Center, such as according to practical in warehouse
Carrying situation, the position etc. of robot issues desired velocity amplitude as speed command to the speed of robot picking.Speed
Value can indicate opposite direction with positive and negative.It include a pair of driven universal such as shown in Fig. 4, in every Train Wheel of transfer robot
Wheel, a driving wheel, by respective motor driven.Two motors respectively execute speed command, become two independent speed closed loops
System.Therefore, two independent Speed closed-link systems can execute different speed commands respectively, and then realization turning etc. is dynamic
Make.
Nonlinear neural network is, for example, NARX (Nonlinear Autoregressive Network with
Exogenous Inputs, non-linear active autoregression) neural network, the hidden neuron and delay order of NARX neural network
It can be configured according to actual needs, such as be set as 8 according to test discovery hidden neuron, delay order is set as 2
When can achieve more satisfied effect for the speed control of robot.Time delay order indicates for 2 by the speed of current t moment
Instruction inputs NARX neural network model with the instruction of the historical speed at t-1 moment and t-2 moment and corresponding velocity amplitude.
Step S304, erection rate corresponding with the present speed instruction instruction that nonlinear neural network model is exported are defeated
Enter PID controller.
Since nonlinear neural network model can simulate the speed command and process of input PID controller under varying environment
The non-linear relation of velocity amplitude after PID controller control, therefore, it would be desirable to can reach after PID controller controls
Velocity amplitude unbalanced input neural network model, then the speed command of available corresponding input PID controller, i.e. input are worked as
Preceding speed command obtains corresponding erection rate instruction.For example, in the prior art only by PID controller to the speed of robot
It is controlled, when the speed command for inputting PID controller is 5m/s, the velocity amplitude after PID controller controls is 4m/s,
There are deviations, if it is desired to obtain the actual speed of 5m/s, it may be necessary to speed command be no longer 5m/s but 6m/s, it is non-thread
Nerve network model is then to be obtained according to the non-linear relation of historical data analog rate instruction and velocity amplitude so that practical speed
The erection rate that angle value is 5m/s instructs.
Step S306, PID controller control speed according to erection rate instruction.
PID controller is instructed and is fed back to erection rate using preset scale parameter, integral parameter, differential parameter
The velocity amplitude of the last moment arrived carries out operation, determines rate controlling amount, and rate controlling amount is inputted executing agency, realization pair
Speed is controlled.The control principle of PID controller belongs to the prior art, and details are not described herein.
Nonlinear neural network model in above-described embodiment can simulate speed command and velocity amplitude under varying environment
Non-linear relation, it is non-thread by inputting the historical speed instruction in present speed instruction and preset time with historical speed value
Modified speed command can be obtained in nerve network model, and modified speed command is according to speed command and velocity amplitude
Non-linear relation is modified present speed instruction, then by the modified speed command input based on PID control
Desired velocity amplitude can be then obtained in system, improve the accuracy and precision of speed control.
It also needs to be trained nonlinear neural network model before using nonlinear neural network model and determines it
In parameter, be described below with reference to Fig. 5.
Fig. 5 is the flow chart of another embodiment of method for control speed of the present invention.As shown in figure 5, before step S302
Can also include:
Step S502, the historical speed instruction acquired under historical speed instruction and different operating conditions under different operating conditions correspond to
Historical speed value as training data.
Different operating conditions are, for example, the environmental impact factors such as different loads situation, different temperatures, different runing times.According to non-
The historical data of acquisition is divided into different training data groups by the delay order of linear neural network model.
Step S504, is normalized training data.
Preferably, by the difference of minimum value in the training data and training data before normalization divided by the training number after normalization
According to the difference with minimum value after default normalization as the first ratio；By maxima and minima in the training data before normalization
Difference is divided by the difference of maxima and minima after default normalization as the second ratio；It is equal with the second ratio according to the first ratio,
Training data after determining normalization.Normalized training data can be specifically calculated using the following equation:
Wherein, ymaxAnd yminThe maximum value and minimum value of training data, usually take 1 and -1 respectively after respectively normalizing,
xmaxAnd xminThe maximum value and minimum value of training data before normalizing, y are the training data after normalization.
Step S506 determines the parameter of nonlinear neural network model using normalized training data, so that it is determined that non-
Linear neural network model.
Nonlinear neural network model is, for example, NARX neural network, and network can have three layers altogether, respectively input layer, hidden
Layer and output layer.Input layer number can be historical speed instruction, historical speed value and subsequent time desired speed value
(i.e. present speed instruction).Hidden neuron number obtains being 8 by experience or experiment.Output layer number of nodes can be 1,
Represent the speed command that current time should provide.The order that is delayed can be 2.
Training stage other than training data being normalized, can also arrange at random each group training data
Sequence can increase the knowledge quantity of nonlinear neural network study in this way and improve the recognition capability to the following new data.Then will
Training data unbalanced input neural network model, is trained it because being pre-processed to training data, have compared with
High learning efficiency and training effect.Training process mainly obtains the parameter of nonlinear neural network model, parameter for example including
The corresponding weight of neuron and deviation etc. determine that these parameters have then determined nonlinear neural network model later.
The historical data that above-described embodiment acquires different operating conditions carries out nonlinear neural network model as training data
Training, can enable nonlinear neural network model simulate the non-linear relation of speed command and velocity amplitude in a variety of situations,
It further increases in use process for the accuracy of speed control and precision.In addition, being pre-processed to training data, have
There are higher learning efficiency and training effect.
The off-line training process to nonlinear neural network model is disclosed in above-described embodiment, in actual application
Nonlinear neural network model can also be updated in real time.It is described below with reference to Fig. 6.
Fig. 6 is the flow chart of another embodiment of method for control speed of the present invention.As shown in fig. 6, after step S306
Can also include:
Step S602 is right according to feeding back to the velocity amplitude of nonlinear neural network model in real time or every predetermined period
Nonlinear neural network model is modified.
Preferably, newly-generated speed command and corresponding velocity amplitude fed back in real time or every predetermined period non-thread
Nerve network model is added in training data using newly-generated speed command and corresponding velocity amplitude as historical data to non-thread
Nerve network model carries out on-line training, to obtain amendment nonlinear neural network model.
Above-described embodiment can carry out real-time adaptive update to nonlinear neural network model according to the data of generation, into
The precision and accuracy of one step raising speed control.
An application examples of the invention is described below with reference to Fig. 7.
As shown in fig. 7, Central Control Center t moment, which issues speed command, inputs NARX neural network model, while history
The speed command and velocity amplitude at t-1 moment and t-2 moment are inputted NARX neural network mould according to delay order by data logger
Type exports modified speed command by the operation of NARX neural network model, and inputs PID controller, if necessary to mould
Type is modified, it is also necessary to is modified according to the velocity amplitude of the last moment of feedback and speed command NARX neural network model
It calculates present speed again afterwards and instructs corresponding modified speed command, while PID controller also receives the velocity amplitude of last moment
Progress operation obtains rate controlling amount and exports to executing agency, and executing agency drives robot ambulation to obtain according to rate controlling amount
Values for actual speed, the values for actual speed can feed back to NARX neural network model and carry out the amendment of model, while feed back to PID control
Device processed is used for the speed control of subsequent time.
The present invention also provides a kind of speed control units, are described below with reference to Fig. 8.
Fig. 8 is the flow chart of speed control unit one embodiment of the present invention.As shown in figure 8, the device 80 includes:
Neural network module 802, when the historical speed for instructing present speed, in preset time instructs and is default
Interior historical speed instructs corresponding historical speed value unbalanced input neural network model.
Nonlinear neural network is, for example, NARX neural network；The hidden neuron of NARX neural network can be 8, prolong
When order can be 2.
Pid control module 804, the amendment corresponding with present speed instruction for exporting nonlinear neural network model
Speed command inputs proportional integral differential PID controller, so that PID controller controls speed according to erection rate instruction
System.
In one embodiment, neural network module 802, be also used to acquire under different operating conditions historical speed instruction and
Historical speed under different operating conditions instructs corresponding historical speed value as training data, and training data is normalized, benefit
The parameter of nonlinear neural network model is determined with normalized training data, so that it is determined that nonlinear neural network model.
Preferably, neural network module 802, the difference for minimum value in the training data and training data before normalizing
Divided by the difference of the training data after normalization and minimum value after default normalization as the first ratio, by the training number before normalization
According to the difference of middle maxima and minima divided by the difference of maxima and minima after default normalization as the second ratio, according to first
Ratio is equal with the second ratio, the training data after determining normalization.
In one embodiment, neural network module 802 are also used to non-according to feeding back in real time or every predetermined period
The speed command and velocity amplitude of linear neural network model, are modified nonlinear neural network model.
In one embodiment, pid control module 804, for being joined using preset scale parameter, integral parameter, differential
The velocity amplitude for the last moment that several pairs of erection rate instructions and feedback obtain carries out operation, rate controlling amount is determined, so as to right
Speed is controlled.
The present invention also provides a kind of computer readable storage mediums, are stored thereon with computer program, which is characterized in that should
The step of method for control speed in any one aforementioned embodiment is realized when program is executed by processor.
Those skilled in the art should be understood that the embodiment of the present invention can provide as method, system or computer journey
Sequence product.Therefore, complete hardware embodiment, complete software embodiment or combining software and hardware aspects can be used in the present invention
The form of embodiment.Moreover, it wherein includes the calculating of computer usable program code that the present invention, which can be used in one or more,
Machine can use the meter implemented in non-transient storage medium (including but not limited to magnetic disk storage, CD-ROM, optical memory etc.)
The form of calculation machine program product.
The present invention be referring to according to the method for the embodiment of the present invention, the process of equipment (system) and computer program product
Figure and/or block diagram describe.It is interpreted as to be realized by computer program instructions each in flowchart and/or the block diagram
The combination of process and/or box in process and/or box and flowchart and/or the block diagram.It can provide these computer journeys
Sequence instruct to general purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices processor with
A machine is generated, so that the instruction generation executed by computer or the processor of other programmable data processing devices is used for
Realize the dress for the function of specifying in one or more flows of the flowchart and/or one or more blocks of the block diagram
It sets.
These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works, so that it includes referring to that instruction stored in the computer readable memory, which generates,
Enable the manufacture of device, the command device realize in one box of one or more flows of the flowchart and/or block diagram or
The function of being specified in multiple boxes.
These computer program instructions also can be loaded onto a computer or other programmable data processing device, so that counting
Series of operation steps are executed on calculation machine or other programmable devices to generate computer implemented processing, thus in computer or
The instruction executed on other programmable devices is provided for realizing in one or more flows of the flowchart and/or block diagram one
The step of function of being specified in a box or multiple boxes.
The foregoing is merely presently preferred embodiments of the present invention, is not intended to limit the invention, it is all in spirit of the invention and
Within principle, any modification, equivalent replacement, improvement and so on be should all be included in the protection scope of the present invention.
Claims (14)
1. a kind of method for control speed characterized by comprising
By the historical speed instruction in present speed instruction, preset time and the historical speed instruction pair in the preset time
The historical speed value unbalanced input neural network model answered；
The erection rate instruction input ratio corresponding with present speed instruction that the nonlinear neural network model is exported
Example integral differential PID controller, so that PID controller controls speed according to erection rate instruction.
2. the method according to claim 1, wherein further include:
It acquires the historical speed instruction under different operating conditions and the historical speed under the different operating conditions instructs corresponding history speed
Angle value is as training data；
The training data is normalized；
The parameter of the nonlinear neural network model is determined using normalized training data, so that it is determined that the non-linear mind
Through network model.
3. according to the method described in claim 2, it is characterized in that, the described training data is normalized includes:
The difference of minimum value in training data and training data before normalization is returned divided by the training data after normalization with default
The difference of minimum value is as the first ratio after one change；
By the difference of maxima and minima in the training data before normalization divided by maxima and minima after default normalization
Difference is used as the second ratio；
Pass through the training data after first ratio determination equal with second ratio is normalized.
4. the method according to claim 1, wherein further include:
It is right according to feeding back to the speed command and velocity amplitude of the nonlinear neural network model in real time or every predetermined period
The nonlinear neural network model is modified.
5. the method according to claim 1, wherein
The PID controller instruct erection rate using preset scale parameter, integral parameter, differential parameter and instead
The velocity amplitude for presenting obtained last moment carries out operation, rate controlling amount is determined, to control speed.
6. method according to claim 1-5, which is characterized in that
The nonlinear neural network is non-linear active autoregression NARX neural network, the hidden layer mind of the NARX neural network
It is 8 through member, delay order is 2.
7. a kind of speed control unit characterized by comprising
Neural network module, historical speed instruction and the preset time for instructing present speed, in preset time
Interior historical speed instructs corresponding historical speed value unbalanced input neural network model；
Pid control module corresponding with present speed instruction is repaired for export the nonlinear neural network model
Positive speed command inputs proportional integral differential PID controller, so as to PID controller according to the erection rate instruct to speed into
Row control.
8. device according to claim 7, which is characterized in that
The neural network module is also used to acquire going through under the instruction of the historical speed under different operating conditions and the different operating condition
The corresponding historical speed value of history speed command is normalized the training data, after normalization as training data
Training data determine the parameter of the nonlinear neural network model, so that it is determined that the nonlinear neural network model.
9. device according to claim 8, which is characterized in that
The difference of minimum value in training data and training data before normalization is returned divided by the training data after normalization with default
The difference of minimum value is as the first ratio after one change, by the difference of maxima and minima in the training data before normalization divided by default
The difference of maxima and minima is as the second ratio after normalization, by making first ratio equal with second ratio
To determine the training data after normalization.
10. device according to claim 7, which is characterized in that
The neural network module is also used to basis and feeds back to the nonlinear neural network mould in real time or every predetermined period
The speed command and velocity amplitude of type are modified the nonlinear neural network model.
11. device according to claim 7, which is characterized in that
The pid control module also using preset scale parameter, integral parameter, differential parameter to the erection rate instruct with
And feed back the velocity amplitude of obtained last moment and carry out operation, rate controlling amount is determined, to control speed.
12. according to the described in any item devices of claim 7-11, which is characterized in that
The nonlinear neural network is non-linear active autoregression NARX neural network, the hidden layer mind of the NARX neural network
It is 8 through member, delay order is 2.
13. a kind of speed control unit characterized by comprising
Memory；And
It is coupled to the processor of the memory, the processor is configured to the instruction based on storage in the memory,
Execute method for control speed as claimed in any one of claims 1 to 6.
14. a kind of computer readable storage medium, is stored thereon with computer program, which is characterized in that the program is by processor
The step of any one of claim 1-6 the method is realized when execution.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710167446.7A CN106877746B (en) | 2017-03-21 | 2017-03-21 | Method for control speed and speed control unit |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710167446.7A CN106877746B (en) | 2017-03-21 | 2017-03-21 | Method for control speed and speed control unit |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106877746A CN106877746A (en) | 2017-06-20 |
| CN106877746B true CN106877746B (en) | 2019-01-01 |
Family
ID=59173267
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710167446.7A Active CN106877746B (en) | 2017-03-21 | 2017-03-21 | Method for control speed and speed control unit |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106877746B (en) |
Families Citing this family (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109343343A (en) * | 2018-09-19 | 2019-02-15 | 广州大学 | Steering gear control system and method based on deep learning and deep neural network |
| CN109159134A (en) * | 2018-10-09 | 2019-01-08 | 上海思依暄机器人科技股份有限公司 | A kind of robot control method and robot |
| CN109581874B (en) * | 2018-12-29 | 2022-04-05 | 百度在线网络技术（北京）有限公司 | Method and apparatus for generating information |
| CN109951123B (en) * | 2019-04-02 | 2020-10-30 | 江南大学 | Speed control method for linear induction motor |
| CN110518851A (en) * | 2019-07-19 | 2019-11-29 | 杭州洲钜电子科技有限公司 | Compressor control method, system and storage medium neural network based |
| CN110843793B (en) * | 2019-12-04 | 2021-11-23 | 苏州智加科技有限公司 | Longitudinal control system and method of automatic driving vehicle based on feedforward control |
| CN111267109B (en) * | 2020-03-24 | 2021-07-02 | 华中科技大学 | Robot speed planning method and system based on reinforcement learning |
| CN111721556B (en) * | 2020-06-29 | 2022-05-17 | 中车长春轨道客车股份有限公司 | Impact speed control method and device, storage medium and electronic equipment |
| CN112688615B (en) * | 2020-12-25 | 2022-06-17 | 绍兴市上虞区武汉理工大学高等研究院 | Alternating current servo driving method and device based on artificial intelligence |
| CN112849899B (en) * | 2020-12-29 | 2022-10-21 | 深圳市海柔创新科技有限公司 | Storage management method, device, equipment, medium, program product and system |
| CN115149873B (en) * | 2022-09-05 | 2022-12-02 | 中科航迈数控软件(深圳)有限公司 | Servo motor set control method based on neural network model and related equipment |
| CN115607193B (en) * | 2022-11-09 | 2023-07-28 | 深圳英美达医疗技术有限公司 | Driving control method, device and equipment for intravascular ultrasonic optical imaging probe |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2003256005A (en) * | 2002-03-01 | 2003-09-10 | Omron Corp | Control method and control device |
| CN102045021A (en) * | 2009-10-12 | 2011-05-04 | 通用汽车环球科技运作公司 | Methods, systems and apparatus for dynamically controlling an electric motor that drives an oil pump |
| CN103151980A (en) * | 2012-12-29 | 2013-06-12 | 江苏大学 | Brushless direct current motor controller for automotive electric power storage (EPS) and realizing method thereof |
| CN103888044A (en) * | 2014-02-25 | 2014-06-25 | 江苏大学 | Parameter self-tuning method for fuzzy PID controller |
| CN104836498A (en) * | 2015-04-21 | 2015-08-12 | 苏州德锐朗智能科技有限公司 | Auto-tuning PID generator control system based on artificial neural network |
-
2017
- 2017-03-21 CN CN201710167446.7A patent/CN106877746B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP2003256005A (en) * | 2002-03-01 | 2003-09-10 | Omron Corp | Control method and control device |
| CN102045021A (en) * | 2009-10-12 | 2011-05-04 | 通用汽车环球科技运作公司 | Methods, systems and apparatus for dynamically controlling an electric motor that drives an oil pump |
| CN103151980A (en) * | 2012-12-29 | 2013-06-12 | 江苏大学 | Brushless direct current motor controller for automotive electric power storage (EPS) and realizing method thereof |
| CN103888044A (en) * | 2014-02-25 | 2014-06-25 | 江苏大学 | Parameter self-tuning method for fuzzy PID controller |
| CN104836498A (en) * | 2015-04-21 | 2015-08-12 | 苏州德锐朗智能科技有限公司 | Auto-tuning PID generator control system based on artificial neural network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106877746A (en) | 2017-06-20 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106877746B (en) | Method for control speed and speed control unit | |
| CN107160398B (en) | The safe and reliable control method of Rigid Robot Manipulator is limited based on the total state for determining study | |
| Nedic et al. | Optimal cascade hydraulic control for a parallel robot platform by PSO | |
| Wen et al. | Elman fuzzy adaptive control for obstacle avoidance of mobile robots using hybrid force/position incorporation | |
| CN105955206B (en) | A kind of Multi-axis motion control method based on data-driven and parameter hybrid optimization | |
| CN108133258A (en) | A kind of mixing global optimization method | |
| CN110134165B (en) | Reinforced learning method and system for environmental monitoring and control | |
| CN103927451B (en) | Ontrack identification method for system parameters of space robot | |
| CN104493826A (en) | Neural network control-based reload robot position controller | |
| CN103616817B (en) | A kind of servo turnable controller based on switched nonlinear systems | |
| Yu et al. | Stable neural PID anti-swing control for an overhead crane | |
| Ali et al. | Combined ANFIS method with FA, PSO, and ICA as Steering control optimization on electric car | |
| CN105022269A (en) | Method and device for controlling bionic robotic fish joint | |
| CN106773688A (en) | A kind of direct adaptive control method and device | |
| CN108828959A (en) | A kind of novel bridge crane is anti-sway with position control method and device | |
| Moness et al. | Real-time Mamdani-like fuzzy and fusion-based fuzzy controllers for balancing two-wheeled inverted pendulum | |
| Zhang et al. | A novel finite-time-gain-adjustment controller design method for UAVs tracking time-varying targets | |
| CN112462608B (en) | Discrete sliding mode track and speed tracking control method for high-speed train | |
| Chen et al. | An overview of robust reinforcement learning | |
| CN108693776A (en) | A kind of robust control method of Three Degree Of Freedom Delta parallel robots | |
| Li et al. | Improved Q-learning based route planning method for UAVs in unknown environment | |
| Abdulla et al. | Roll control system design using auto tuning LQR technique | |
| Shojaei et al. | A hybrid control scheme for a rotational inverted pendulum | |
| Kaur et al. | Position control of overhead cranes using fuzzy controller | |
| CN109176529B (en) | Self-adaptive fuzzy control method for coordinated movement of space 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 | ||
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20190730 Address after: 300 457 days Tianjin Binhai New Area, Tianjin Economic and Technological Development Zone, No. 10 Fourth Avenue, Phase 2, First Floor Patentee after: Tianjin Jingdong Shentuo Robot Technology Co., Ltd. Address before: East Building 11, 100195 Beijing city Haidian District xingshikou Road No. 65 west Shan creative garden district 1-4 four layer of 1-4 layer Co-patentee before: Beijing Jingdong Century Commerce Co., Ltd. Patentee before: Beijing Jingdong Shangke Information Technology Co., Ltd. |
|
| TR01 | Transfer of patent right | ||
| TR01 | Transfer of patent right |
Effective date of registration: 20200521 Address after: Room A1905, 19th floor, No. 2 Building, 18 Kechuang 11th Street, Beijing Daxing District, Beijing Patentee after: Beijing Jingdong Qianshi Technology Co.,Ltd. Address before: 300 457 days Tianjin Binhai New Area, Tianjin Economic and Technological Development Zone, No. 10 Fourth Avenue, Phase 2, First Floor Patentee before: Tianjin Jingdong Shentuo Robot Technology Co.,Ltd. |