CN106054893B - The control system and method for intelligent vehicle - Google Patents
The control system and method for intelligent vehicle Download PDFInfo
- Publication number
- CN106054893B CN106054893B CN201610510460.8A CN201610510460A CN106054893B CN 106054893 B CN106054893 B CN 106054893B CN 201610510460 A CN201610510460 A CN 201610510460A CN 106054893 B CN106054893 B CN 106054893B
- Authority
- CN
- China
- Prior art keywords
- intelligent vehicle
- neural network
- dynamic
- dynamic fuzzy
- parameter
- 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.)
- Expired - Fee Related
Links
- 238000000034 method Methods 0.000 title claims abstract description 24
- 238000013528 artificial neural network Methods 0.000 claims abstract description 97
- 238000001514 detection method Methods 0.000 claims abstract description 18
- 238000013138 pruning Methods 0.000 claims abstract description 18
- 230000004888 barrier function Effects 0.000 claims description 12
- 238000004891 communication Methods 0.000 claims description 10
- 108010074506 Transfer Factor Proteins 0.000 claims description 6
- 241000208340 Araliaceae Species 0.000 claims description 4
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 claims description 4
- 235000003140 Panax quinquefolius Nutrition 0.000 claims description 4
- 235000008434 ginseng Nutrition 0.000 claims description 4
- 230000003321 amplification Effects 0.000 claims description 3
- 238000003199 nucleic acid amplification method Methods 0.000 claims description 3
- 230000001537 neural effect Effects 0.000 claims description 2
- 210000004218 nerve net Anatomy 0.000 claims 4
- 230000001276 controlling effect Effects 0.000 description 8
- 238000012937 correction Methods 0.000 description 4
- 230000008901 benefit Effects 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 230000001105 regulatory effect Effects 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000003708 edge detection Methods 0.000 description 1
- 230000007717 exclusion Effects 0.000 description 1
- 238000011478 gradient descent method Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 238000010295 mobile communication Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000000465 moulding Methods 0.000 description 1
- 210000005036 nerve Anatomy 0.000 description 1
- 238000010606 normalization Methods 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000002604 ultrasonography Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0231—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means
- G05D1/0246—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means using a video camera in combination with image processing means
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0212—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory
- G05D1/0221—Control of position or course in two dimensions specially adapted to land vehicles with means for defining a desired trajectory involving a learning process
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course, altitude or attitude of land, water, air or space vehicles, e.g. using automatic pilots
- G05D1/02—Control of position or course in two dimensions
- G05D1/021—Control of position or course in two dimensions specially adapted to land vehicles
- G05D1/0255—Control of position or course in two dimensions specially adapted to land vehicles using acoustic signals, e.g. ultra-sonic singals
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Radar, Positioning & Navigation (AREA)
- Remote Sensing (AREA)
- Aviation & Aerospace Engineering (AREA)
- General Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Multimedia (AREA)
- Electromagnetism (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Acoustics & Sound (AREA)
- Traffic Control Systems (AREA)
- Control Of Driving Devices And Active Controlling Of Vehicle (AREA)
- Steering Control In Accordance With Driving Conditions (AREA)
Abstract
The invention discloses a kind of control system of intelligent vehicle and methods, belong to technical field of vehicle.The control system includes: detection module, for obtaining the driving information of intelligent vehicle；Control module, for according to the driving information, determine input parameter, and the dynamic fuzzy neural network for using Pruning Algorithm to delete node is inputted by parameter is inputted, obtain pid control parameter, the input parameter include the position of the intelligent vehicle opposite lane line, the spacing of the intelligent vehicle and front obstacle, the travel speed of the intelligent vehicle, the first difference of the travel speed, the intelligent vehicle rudder angle；Execution module, for controlling the speed and steering angle of intelligent vehicle according to pid control parameter.The present invention can realize the automatic steady operation of intelligent vehicle, strong robustness in different environments.And the redundant node in neural network is deleted using Pruning Algorithm, the operational efficiency of neural network can be greatly improved.
Description
Technical field
The present invention relates to technical field of vehicle, in particular to the control system and method for a kind of intelligent vehicle.
Background technique
Intelligent vehicle automatically can manipulate and drive vehicle and get around barrier and advance along scheduled road, have road roadblock
Hinder automatic identification, automatic alarm, automatic braking, automatically keep the functions such as safe distance, speed and cruise control, intelligent vehicle is this
Advantages characteristic is extremely widely applied so that it has in terms of military affairs, road engineering, danger souding and dangerous situation exclusion.Mesh
Before, intelligent vehicle has become the research hotspot of automotive field.
The control algolithm of intelligent vehicle generallys use fuzzy neural network realization, the node of each layer of fuzzy neural network at this stage
Quantity is difficult to determine, after establishing neural network, each node layer quantity can not be changed, and all can in the neural network typically set up
There are redundant nodes.Since the relationship of non-linear direct ratio, fuzzy neural is presented in the calculation amount and number of nodes of fuzzy neural network
Redundant node present in network constrains the efficiency of fuzzy neural network.
Summary of the invention
Lower using Fuzzy Neural-network Control intelligent vehicle efficiency in the prior art in order to solve the problems, such as, the present invention is implemented
Example provides the control system and method for a kind of intelligent vehicle.The technical solution is as follows:
On the one hand, the embodiment of the invention provides a kind of control system of intelligent vehicle, the control system includes:
Detection module, for obtaining the driving information of intelligent vehicle, the driving information includes the road letter of the intelligent vehicle
It ceases, the rudder angle of the travel speed of the barrier around the intelligent vehicle, the intelligent vehicle, the intelligent vehicle；
Control module, the driving information for being got according to the detection module determine input parameter, and by institute
The dynamic fuzzy neural network that node is deleted in input parameter input using Pruning Algorithm is stated, proportional integral derivative PID control is obtained
Parameter, it is described input parameter include the position of the intelligent vehicle opposite lane line, the intelligent vehicle and front obstacle spacing,
The travel speed of the intelligent vehicle, the first difference of the travel speed, the intelligent vehicle rudder angle；
Execution module, for controlling the speed and steering angle of the intelligent vehicle according to the pid control parameter.
Specifically, the detection module includes:
Linear array camera, for acquiring the carriageway image where the intelligent vehicle；
Ultrasonic sensor, for detecting barrier；
Velocity sensor, for detecting the travel speed of the intelligent vehicle；
Angular transducer, for detecting the rudder angle of the intelligent vehicle.
Further, the detection module further include:
Amplifying circuit, the signal amplification for detecting ultrasonic sensor；
Analog-digital converter, for the amplified signal of the amplifying circuit to be converted to digital signal and passes to the control
Molding block.
Specifically, the execution module include steering engine for controlling the steering angle of the intelligent vehicle, it is described for controlling
The motor of the speed of intelligent vehicle and control circuit for controlling the steering engine and the motor.
Preferably, the control system further includes communication module, and the communication module is used to be counted with control computer
According to exchange.
On the other hand, the embodiment of the invention provides a kind of control method of intelligent vehicle, the control method includes:
The driving information of intelligent vehicle is obtained, the driving information includes the road information of the intelligent vehicle, the intelligent vehicle
Around barrier, the travel speed of the intelligent vehicle, the intelligent vehicle rudder angle；
The driving information got according to the detection module determines input parameter, and the input parameter includes institute
State the position of intelligent vehicle opposite lane line, the spacing of the intelligent vehicle and front obstacle, the travel speed of the intelligent vehicle, institute
State the first difference of travel speed, the rudder angle of the intelligent vehicle；
The dynamic fuzzy neural network that the input parameter input is deleted to node using Pruning Algorithm, obtains proportion differential
Integral PID control parameter；
The speed and steering angle of the intelligent vehicle are controlled according to the pid control parameter.
Optionally, the control method further include:
Obtain sample set, the sample set include multiple input samples and it is corresponding with multiple input samples correspondingly it is defeated
Sample out；
Dynamic fuzzy neural network is trained using sample set；
The node in dynamic fuzzy neural network is deleted using Pruning Algorithm.
It is further, described that the node in the dynamic fuzzy neural network is deleted using Pruning Algorithm, comprising:
The error for deleting the dynamic fuzzy neural network of first node is calculated, the first node is the dynamic analog
Paste any one node in neural network；
If the error for deleting the dynamic fuzzy neural network after first node is within the set range, described first is deleted
Node.
It is further, described that dynamic fuzzy neural network is trained using sample set, comprising:
Input sample is inputted into dynamic neural network, obtains the pid control parameter of dynamic neural network output；
The output sample of identical input sample is corresponded in the pid control parameter and sample set of comparison dynamic neural network output
This, calculates the error of dynamic fuzzy neural network；
When the error of dynamic fuzzy neural network is more than given threshold, according in error transfer factor dynamic fuzzy neural network
Parameter；
When the error of dynamic fuzzy neural network is in given threshold, stop the ginseng in adjustment dynamic fuzzy neural network
Number.
Optionally, the control method further include:
Data exchange is carried out with control computer；
Dynamic module neural network is corrected according to the content of data exchange.
Technical solution provided in an embodiment of the present invention has the benefit that
By the way that the traveling of the spacing of the position of intelligent vehicle opposite lane line, intelligent vehicle and front obstacle, intelligent vehicle is fast
Input parameter of the rudder angle of degree, the first difference of travel speed and intelligent vehicle as dynamic fuzzy neural network, and according to dynamic
The travel speed and rudder angle of the pid control parameter control intelligent vehicle of state fuzzy neural network output, may be implemented oneself of intelligent vehicle
Dynamic even running, strong robustness and control precision height.In addition, by using Pruning Algorithm to superfluous in dynamic fuzzy neural network
Remaining node is deleted, and the operational efficiency of dynamic fuzzy neural network can be greatly improved.
Detailed description of the invention
To describe the technical solutions in the embodiments of the present invention more clearly, make required in being described below to embodiment
Attached drawing is briefly described, it should be apparent that, drawings in the following description are only some embodiments of the invention, for
For those of ordinary skill in the art, without creative efforts, it can also be obtained according to these attached drawings other
Attached drawing.
Fig. 1 is a kind of structural schematic diagram of the control system of intelligent vehicle provided in an embodiment of the present invention；
Fig. 2 is a kind of flow chart of the control method of intelligent vehicle provided in an embodiment of the present invention.
Specific embodiment
To make the object, technical solutions and advantages of the present invention clearer, below in conjunction with attached drawing to embodiment party of the present invention
Formula is described in further detail.
The embodiment of the invention provides a kind of control systems of intelligent vehicle, are adapted to carry out the intelligent vehicle of the offer of embodiment one
Control method, referring to Fig. 1, which includes: detection module 101, control module 102 and execution module 103.Control mould
Block 102 is connect with detection module 101 and execution module 103 respectively.
Wherein, detection module 101 is used to obtain the driving information of intelligent vehicle, which includes but is not limited to intelligent vehicle
Road information, the barrier around intelligent vehicle, the travel speed of intelligent vehicle, intelligent vehicle rudder angle.Control module 102 is used for root
According to the driving information that detection module 101 is got, input parameter is determined, and input parameter input is deleted into section using Pruning Algorithm
The dynamic fuzzy neural network of point, obtains (proportion integral derivative, abbreviation PID) control parameter, should
Input parameter include intelligent vehicle opposite lane line position (may include distance apart from lane line, the angle with lane line),
The spacing of intelligent vehicle and front obstacle, the travel speed of intelligent vehicle, the first difference of travel speed, intelligent vehicle rudder angle.It holds
The speed and steering angle for the pid control parameter control intelligent vehicle that row module 103 is used to be obtained according to control module 102.
In the present embodiment, detection module 101 may include:
Linear array camera 1011, for acquiring the carriageway image where intelligent vehicle；
Ultrasonic sensor 1012, for detecting barrier；
Velocity sensor 1013, for detecting the travel speed of intelligent vehicle；
Angular transducer 1014, for detecting the rudder angle of intelligent vehicle.
Wherein, linear array camera 1011, ultrasonic sensor 1012, velocity sensor 1013 and angular transducer 1014
It is electrically connected respectively with control module 102.
In the concrete realization, linear array camera 1012 can shoot the lane line image of intelligent vehicle two sides, control module 102
Lane detection is carried out to image, determines intelligent vehicle opposite lane line using image processing algorithm (such as edge detection algorithm etc.)
Position.Ultrasonic sensor 1012 persistently sends ultrasonic wave under the control of control module 102, and detects barrier and be reflected back
Ultrasonic wave, the time difference that control module 102 sends and receives according to ultrasonic wave can determine the spacing of intelligent vehicle and barrier.Speed
Degree sensor 1013 can directly detect the travel speed of intelligent vehicle, and be transferred to control module 102.Angular transducer 1014 can
With the angular transducer included using steering engine on intelligent vehicle.
Detection module 101 can also include amplifying circuit 1014 and analog-digital converter 1015, amplifying circuit and 1014 and mould
Number converter 1015 is connected between ultrasonic sensor 1012 and control module 102, in order to which control module 102 reads ultrasound
The output signal of wave sensor 1012.Specifically, amplifying circuit 1014 is used for the signal for detecting ultrasonic sensor 1012
Amplification, analog-digital converter 1015 is for being converted to digital signal for the amplified signal of amplifying circuit 1014 and passing to control mould
Block 102.
Specifically, control module 102 may include embedded microprocessor, and degree of integration is high, low in cost, environment adapts to
Property is good, is widely used.
Optionally, execution module 103 may include steering engine 1031 for controlling the steering angle of intelligent vehicle, for controlling intelligence
Can vehicle speed motor 1032 and control circuit 1033 for controlling steering engine 1031 and motor 1032.Specifically,
Control circuit 1033 may include PID controller.
In the concrete realization, execution module 103 can also include regulated power supply 1034, regulated power supply 1034 and control circuit
1033 connections to realize control of the control circuit 1033 to 1032 revolving speed of motor, and then adjust the travel speed of intelligent vehicle.
Optionally, which can also include:
Communication module 104, for carrying out data exchange with control computer.
Specifically, the content of data exchange may include the lane information where intelligent vehicle, such as lane shape (such as it is straight
Line lane, left bend, right bend etc.), lane width, the information such as turning radius, then control module 102 is specifically used for according to intelligence
The carriageway image where lane information and intelligent vehicle where vehicle, determines the position of intelligent vehicle opposite lane line, so that dynamic
Fuzzy neural network can adapt to different environment.The content of data exchange can also include the amendment of dynamic fuzzy neural network
Instruction, which may include the correction value of the parameter in dynamic fuzzy neural network；Then control module 102 is also used to root
According to the parameter in correction value adjustment dynamic fuzzy neural network, so as to realize the on-line tuning to vehicle control algorithms.
When realization, communication module 104 can be wire communication module, or wireless communication module, radio communication mold
Block includes but is not limited to mobile communication module, Wireless Fidelity (Wireless Fidelity, abbreviation WIFI) module etc..
The embodiment of the present invention is by by the spacing of the position of intelligent vehicle opposite lane line, intelligent vehicle and front obstacle, intelligence
Can the travel speed of vehicle, the first difference of travel speed and intelligent vehicle input of the rudder angle as dynamic fuzzy neural network
Parameter, and the travel speed and rudder angle of the pid control parameter control intelligent vehicle according to dynamic fuzzy neural network output, Ke Yishi
The automatic steady operation of existing intelligent vehicle, strong robustness and control precision height.In addition, by using Pruning Algorithm to dynamic fuzzy mind
It is deleted through the redundant node in network, the operational efficiency of dynamic fuzzy neural network can be greatly improved.In addition, the control
Each section of system is all made of common apparatus realization, and easy, practical, using flexible, cost of implementation are low.
The embodiment of the invention also provides a kind of control method of intelligent vehicle, which can use shown in FIG. 1
Control system realization, referring to fig. 2, this method comprises:
Step 201: obtaining sample set.
In the present embodiment, sample set includes multiple input samples and one-to-one output corresponding with multiple input samples
Sample.Each input sample includes one group of input parameter, and every group of input parameter includes: the position of intelligent vehicle opposite lane line, intelligence
Can the spacing of vehicle and front obstacle, the travel speed of intelligent vehicle, the first difference of travel speed, intelligent vehicle rudder angle.Each
Exporting sample includes one group of pid control parameter.
Step 202: dynamic fuzzy neural network being trained using sample set.
Specifically, dynamic fuzzy neural network is trained i.e. using sample set to dynamic fuzzy nerve using sample set
Parameter in network is adjusted.
Specifically, dynamic fuzzy neural network generally includes input layer, blurring layer, rules layer, normalization layer, recurrence layer
And output layer.Wherein, each layer includes multiple nodes, is configured with a parameter between any two node of adjacent two layers, i.e.,
Parameter in dynamic fuzzy neural network.
Optionally, which may include:
Input sample is inputted into dynamic neural network, obtains the pid control parameter of dynamic neural network output；
The output sample of identical input sample is corresponded in the pid control parameter and sample set of comparison dynamic neural network output
This, calculates the error of dynamic fuzzy neural network；
When the error of dynamic fuzzy neural network is more than given threshold, according in error transfer factor dynamic fuzzy neural network
Parameter；
When the error of dynamic fuzzy neural network is in given threshold, stop the ginseng in adjustment dynamic fuzzy neural network
Number.
Specifically, identical input sample is corresponded in the pid control parameter and sample set of the output of comparison dynamic neural network
Output sample, calculate the error of dynamic fuzzy neural network, may include:
The error of dynamic fuzzy neural network is calculated according to following formula:
E=| x-a |/a；
Wherein, E is error, and x is the pid control parameter of dynamic neural network output, and a is identical defeated to correspond in sample set
Enter the output sample of sample, | | absolute value is sought in expression.
Specifically, according to the parameter in error transfer factor dynamic fuzzy neural network, may include:
Using gradient descent method according to the parameter in error transfer factor dynamic fuzzy neural network.
By constantly adjusting the parameter of neural network according to output result, gradually reduce between output result and ideal value
Gap improves the accuracy of neural network output result.
Step 203: the node in dynamic fuzzy neural network being deleted using Pruning Algorithm.
Optionally, which may include:
The error for deleting the dynamic fuzzy neural network of first node is calculated, first node is in dynamic fuzzy neural network
Any one node；
If the error for deleting dynamic fuzzy neural network after first node is within the set range, the first node is deleted.
It should be noted that being greatly improved by deleting the acceptable node of error in the case where guaranteeing computational accuracy
Computational efficiency.
It should be noted that step 201- step 203 is optional step, it is used to form and node is deleted using Pruning Algorithm
Dynamic fuzzy neural network.
Step 204: obtaining the driving information of intelligent vehicle.
In the present embodiment, which includes but is not limited to carriageway image where intelligent vehicle, around intelligent vehicle
The rudder angle of barrier, the travel speed of intelligent vehicle and intelligent vehicle.
The step 204 may include:
The driving information of vehicle is obtained every setting time, to adjust the travel speed of intelligent vehicle according to the present situation in time
The direction and.
When realization, which can be realized using the detection module in Fig. 1.
Step 205: according to driving information, determine input parameter, input parameter include intelligent vehicle opposite lane line position,
The spacing of intelligent vehicle and front obstacle, the travel speed of intelligent vehicle, the first difference of the travel speed, intelligent vehicle rudder
Angle.
Step 206: parameter input will be inputted, Pruning Algorithm is used to delete the dynamic fuzzy neural network of node, obtain PID
Control parameter.
Step 207: the speed and steering angle of intelligent vehicle are controlled according to pid control parameter.
In the concrete realization, the traveling control of intelligent vehicle can be realized using PID controller.Due to being mainly in low-frequency range
PI control is worked, and steady-state error can be eliminated or reduce；It is mainly that PD works in medium-high frequency section, response speed can be improved
Degree, comprehensively improves system control performance.
The step 205~step 207 can be realized using the control module in Fig. 1.
Optionally, which can also include:
Data exchange is carried out with control computer.
The step can be realized by the communication module in Fig. 1.
Specifically, the content of data exchange may include the lane information where intelligent vehicle, such as lane shape (such as it is straight
Line lane, left bend, right bend etc.), lane width, the information such as turning radius, then the position of aforementioned intelligent vehicle opposite lane line
It can be determined according to the carriageway image where the lane information and intelligent vehicle where intelligent vehicle, so that dynamic fuzzy neural network
It can adapt to different environment.
The content of data exchange can also include the revision directive of dynamic fuzzy neural network, which may include
The correction value of parameter in dynamic fuzzy neural network；Then control method further include: dynamic fuzzy mind is adjusted according to correction value
Through the parameter in network, so as to realize the on-line tuning to vehicle control algorithms.
The embodiment of the present invention is by by the spacing of the position of intelligent vehicle opposite lane line, intelligent vehicle and front obstacle, intelligence
Can the travel speed of vehicle, the first difference of travel speed and intelligent vehicle input of the rudder angle as dynamic fuzzy neural network
Parameter, and the travel speed and rudder angle of the pid control parameter control intelligent vehicle according to dynamic fuzzy neural network output, Ke Yishi
The automatic steady operation of existing intelligent vehicle, strong robustness and control precision height.In addition, by using Pruning Algorithm to dynamic fuzzy mind
It is deleted through the redundant node in network, the operational efficiency of dynamic fuzzy neural network can be greatly improved.
It should be understood that the control system of intelligent vehicle provided by the above embodiment is when controlling intelligent vehicle, only with above-mentioned
The division progress of each functional unit can according to need and for example, in practical application by above-mentioned function distribution by different
Functional unit is completed, i.e., the internal structure of system is divided into different functional units, with complete it is described above whole or
Partial function.In addition, the control system of intelligent vehicle provided by the above embodiment and the control method embodiment of intelligent vehicle belong to together
One design, specific implementation process are detailed in embodiment of the method, and which is not described herein again.
Those of ordinary skill in the art will appreciate that realizing that all or part of the steps of above-described embodiment can pass through hardware
It completes, relevant hardware can also be instructed to complete by program, the program can store in a kind of computer-readable
In storage medium, storage medium mentioned above can be read-only memory, disk or CD etc..
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 (8)
1. a kind of control system of intelligent vehicle, which is characterized in that the control system includes:
Detection module, for obtaining the driving information of intelligent vehicle, the driving information includes the road information of the intelligent vehicle, institute
State the rudder angle of barrier around intelligent vehicle, the travel speed of the intelligent vehicle, the intelligent vehicle；
Control module, the driving information for being got according to the detection module determine input parameter, and will be described defeated
Enter the dynamic fuzzy neural network that node is deleted in parameter input using Pruning Algorithm, obtain proportional integral derivative pid control parameter,
The input parameter includes the spacing, described of the position of the intelligent vehicle opposite lane line, the intelligent vehicle and front obstacle
The travel speed of intelligent vehicle, the first difference of the travel speed, the intelligent vehicle rudder angle,
The control module is also used to；
Obtain sample set, the sample set include multiple input samples and it is corresponding with the multiple input sample correspondingly it is defeated
Sample out；
The dynamic fuzzy neural network is trained using the sample set, it is described neural to dynamic fuzzy using sample set
Network is trained, comprising:
The input sample is inputted into dynamic neural network, obtains the pid control parameter of the dynamic neural network output,
It compares in the pid control parameter and the sample set of the dynamic neural network output and corresponds to the defeated of identical input sample
Sample out calculates the error of the dynamic fuzzy neural network, wherein calculates the dynamic fuzzy nerve net according to following formula
The error of network:
E=| x-a |/a,
Wherein, E is error, and x is the pid control parameter of dynamic neural network output, and a is to correspond to identical input sample in sample set
This output sample, | | absolute value is sought in expression,
When the error of the dynamic fuzzy neural network is more than given threshold, according to the error transfer factor dynamic fuzzy nerve net
Parameter in network,
When the error of the dynamic fuzzy neural network is in given threshold, stop the ginseng in adjustment dynamic fuzzy neural network
Number；
The node in the dynamic fuzzy neural network is deleted using Pruning Algorithm.
2. control system according to claim 1, which is characterized in that the detection module includes:
Linear array camera, for acquiring the carriageway image where the intelligent vehicle；
Ultrasonic sensor, for detecting barrier；
Velocity sensor, for detecting the travel speed of the intelligent vehicle；
Angular transducer, for detecting the rudder angle of the intelligent vehicle.
3. control system according to claim 2, which is characterized in that the detection module further include:
Amplifying circuit, the signal amplification for detecting ultrasonic sensor；
Analog-digital converter, for the amplified signal of the amplifying circuit to be converted to digital signal and passes to the control mould
Block.
4. control system according to claim 1-3, which is characterized in that the control system further includes executing mould
Block, the execution module include the steering engine for controlling the steering angle of the intelligent vehicle, the speed for controlling the intelligent vehicle
Motor and control circuit for controlling the steering engine and the motor.
5. control system according to claim 1-3, which is characterized in that the control system further includes communication mould
Block, the communication module are used to carry out data exchange with control computer.
6. a kind of control method of intelligent vehicle, which is characterized in that the control method includes:
The driving information of intelligent vehicle is obtained, the driving information includes the road information of the intelligent vehicle, around the intelligent vehicle
Barrier, the travel speed of the intelligent vehicle, the intelligent vehicle rudder angle；
According to the driving information, determine input parameter, the input parameter include the intelligent vehicle opposite lane line position,
It is the spacing of the intelligent vehicle and front obstacle, the travel speed of the intelligent vehicle, the first difference of the travel speed, described
The rudder angle of intelligent vehicle,
The control method further include:
Obtain sample set, the sample set include multiple input samples and it is corresponding with the multiple input sample correspondingly it is defeated
Sample out；
Dynamic fuzzy neural network is trained using the sample set,
It is described that dynamic fuzzy neural network is trained using sample set, comprising:
The input sample is inputted into dynamic neural network, obtains the pid control parameter of the dynamic neural network output,
It compares in the pid control parameter and the sample set of the dynamic neural network output and corresponds to the defeated of identical input sample
Sample out calculates the error of the dynamic fuzzy neural network, wherein calculates the dynamic fuzzy nerve net according to following formula
The error of network:
E=| x-a |/a,
Wherein, E is error, and x is the pid control parameter of dynamic neural network output, and a is to correspond to identical input sample in sample set
This output sample, | | absolute value is sought in expression,
When the error of the dynamic fuzzy neural network is more than given threshold, according to the error transfer factor dynamic fuzzy nerve net
Parameter in network,
When the error of the dynamic fuzzy neural network is in given threshold, stop the ginseng in adjustment dynamic fuzzy neural network
Number,
The node in the dynamic fuzzy neural network is deleted using Pruning Algorithm；
The dynamic fuzzy neural network that the input parameter input is deleted to node using Pruning Algorithm, obtains proportional integral derivative
Pid control parameter；
The speed and steering angle of the intelligent vehicle are controlled according to the pid control parameter.
7. control method according to claim 6, which is characterized in that described refreshing to the dynamic fuzzy using Pruning Algorithm
It is deleted through the node in network, comprising:
The error for deleting the dynamic fuzzy neural network of first node is calculated, the first node is the dynamic fuzzy mind
Through any one node in network；
If the error for deleting the dynamic fuzzy neural network after first node is within the set range, the first segment is deleted
Point.
8. control method according to claim 6 or 7, which is characterized in that the control method further include:
Data exchange is carried out with control computer.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610510460.8A CN106054893B (en) | 2016-06-30 | 2016-06-30 | The control system and method for intelligent vehicle |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610510460.8A CN106054893B (en) | 2016-06-30 | 2016-06-30 | The control system and method for intelligent vehicle |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106054893A CN106054893A (en) | 2016-10-26 |
| CN106054893B true CN106054893B (en) | 2019-07-26 |
Family
ID=57201638
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610510460.8A Expired - Fee Related CN106054893B (en) | 2016-06-30 | 2016-06-30 | The control system and method for intelligent vehicle |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106054893B (en) |
Families Citing this family (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US11315018B2 (en) | 2016-10-21 | 2022-04-26 | Nvidia Corporation | Systems and methods for pruning neural networks for resource efficient inference |
| KR102342143B1 (en) * | 2017-08-08 | 2021-12-23 | 주식회사 만도모빌리티솔루션즈 | Deep learning based self-driving car, deep learning based self-driving control device, and deep learning based self-driving control method |
| CN107992059B (en) * | 2018-01-08 | 2020-04-24 | 陕西理工大学 | Control method of electromagnetic intelligent vehicle |
| CN108319267B (en) * | 2018-01-30 | 2021-07-30 | 南京溧水高新产业股权投资有限公司 | Unmanned taxi |
| CN108196442A (en) * | 2018-03-02 | 2018-06-22 | 广州大学 | Steering gear control system and method based on fuzzy neural PID control and absolute encoder |
| CN108859951B (en) * | 2018-05-18 | 2021-08-03 | 武汉理工大学 | Steering early warning system of automobile steering system |
| US20190361454A1 (en) * | 2018-05-24 | 2019-11-28 | GM Global Technology Operations LLC | Control systems, control methods and controllers for an autonomous vehicle |
| CN109407666A (en) * | 2018-10-11 | 2019-03-01 | 智久(厦门)机器人科技有限公司上海分公司 | A kind of control method of vehicle, device, equipment and storage medium |
| CN109739228B (en) * | 2018-12-28 | 2021-12-10 | 创泽智能机器人集团股份有限公司 | Speed regulation parameter self-training method for security robot |
| CN112651456B (en) * | 2020-12-31 | 2023-08-08 | 遵义师范学院 | Unmanned vehicle control method based on RBF neural network |
| CN113200086A (en) * | 2021-06-21 | 2021-08-03 | 盐城工学院 | Intelligent vehicle steering control system and control method thereof |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101824948A (en) * | 2009-03-05 | 2010-09-08 | 通用汽车环球科技运作公司 | Be used for the adaptive control system that automotive vehicle is used |
| CN102109821A (en) * | 2010-12-30 | 2011-06-29 | 中国科学院自动化研究所 | System and method for controlling adaptive cruise of vehicles |
| CN104532829A (en) * | 2014-12-11 | 2015-04-22 | 贵州新联爆破工程集团有限公司 | Pile foundation steel bar cage manufacturing method |
| CN104914867A (en) * | 2015-06-12 | 2015-09-16 | 吉林大学 | Hexapod robot autonomous navigation closed-loop controller with fuzzy neural network |
-
2016
- 2016-06-30 CN CN201610510460.8A patent/CN106054893B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101824948A (en) * | 2009-03-05 | 2010-09-08 | 通用汽车环球科技运作公司 | Be used for the adaptive control system that automotive vehicle is used |
| CN102109821A (en) * | 2010-12-30 | 2011-06-29 | 中国科学院自动化研究所 | System and method for controlling adaptive cruise of vehicles |
| CN104532829A (en) * | 2014-12-11 | 2015-04-22 | 贵州新联爆破工程集团有限公司 | Pile foundation steel bar cage manufacturing method |
| CN104914867A (en) * | 2015-06-12 | 2015-09-16 | 吉林大学 | Hexapod robot autonomous navigation closed-loop controller with fuzzy neural network |
Non-Patent Citations (2)
| Title |
|---|
| 基于动态剪枝算法的神经网络预测模型;张瀛;《复旦学报》;20150630;第54卷(第3期);正文第313-318页 * |
| 基于多传感器数据融合智能导航车的算法研究;候立梅;《中国优秀硕士学位论文全文数据库》;20140215(第2期);正文第23-34页 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106054893A (en) | 2016-10-26 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106054893B (en) | The control system and method for intelligent vehicle | |
| CN110018632B (en) | Vehicle lane change control method and device | |
| CN108573272A (en) | Track approximating method | |
| CN110221541A (en) | The feed-forward coefficients acquisition device of feedforward controller, method in servo-system | |
| CN108388247A (en) | A kind of unmanned vehicle formation drive manner based on oriented communication network network | |
| CN111679668B (en) | Following control method of networked autonomous fleet based on new time-distance strategy | |
| CN111739342A (en) | Method, device, medium, and vehicle for avoiding vehicle ahead of side | |
| CN112558476A (en) | Non-linear multi-wisdom system leaderless consistency control method based on attack compensation | |
| Flores et al. | Enabling cooperative adaptive cruise control on strings of vehicles with heterogeneous dynamics and powertrains | |
| CN118034284A (en) | Collision prevention control method and device for multiple robots, terminal equipment and storage medium | |
| CN117506907B (en) | Coordinated tracking control method for fractional order multi-mechanical arm system | |
| CN117826590A (en) | Unmanned vehicle formation control method and system based on prepositive following topological structure | |
| CN109686086A (en) | The training of fuzzy control network generates the method and device that speed is suggested at crossing | |
| CN115257736B (en) | Vehicle distance maintenance speed planning method based on fuzzy reasoning true value evolution | |
| CN116991167A (en) | Agricultural machinery linear walking control method, device, terminal and storage medium | |
| CN116107339B (en) | Fault-tolerant cooperative control method for bee colony unmanned aerial vehicle under thrust loss fault | |
| CN109606467A (en) | A kind of vehicle steering method and vehicle | |
| Pledgie et al. | Groups of unmanned vehicles: Differential flatness, trajectory planning, and control | |
| CN115303288A (en) | Vehicle control method, control device and camera device | |
| CN112720478B (en) | Robot torque control method and device, readable storage medium and robot | |
| CN113031644B (en) | Communication time lag oriented aircraft formation control system event triggering method, device and medium | |
| CN113759985A (en) | Unmanned aerial vehicle flight control method, system, device and storage medium | |
| CN114564033B (en) | Multi-rotor unmanned aerial vehicle angle control method and multi-rotor unmanned aerial vehicle | |
| WO2020155020A1 (en) | Environment perception method and device, control method and device, and vehicle | |
| Zhang et al. | Path Prediction Method for Automotive Applications Based on Cubic Spline Interpolation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20190726 |