CN108594799A - The real-time acquisition device of pilotless automobile traffic information and system - Google Patents
The real-time acquisition device of pilotless automobile traffic information and system Download PDFInfo
- Publication number
- CN108594799A CN108594799A CN201810045749.6A CN201810045749A CN108594799A CN 108594799 A CN108594799 A CN 108594799A CN 201810045749 A CN201810045749 A CN 201810045749A CN 108594799 A CN108594799 A CN 108594799A
- Authority
- CN
- China
- Prior art keywords
- data
- time
- real
- road
- pilotless automobile
- 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.)
- Withdrawn
Links
- 238000004891 communication Methods 0.000 claims abstract description 24
- 230000006870 function Effects 0.000 claims description 24
- 238000000034 method Methods 0.000 claims description 24
- 210000002569 neuron Anatomy 0.000 claims description 24
- 238000013528 artificial neural network Methods 0.000 claims description 21
- 238000004422 calculation algorithm Methods 0.000 claims description 13
- 238000004364 calculation method Methods 0.000 claims description 12
- 238000000605 extraction Methods 0.000 claims description 11
- 238000001514 detection method Methods 0.000 claims description 8
- 239000011159 matrix material Substances 0.000 claims description 8
- 210000004218 nerve net Anatomy 0.000 claims description 4
- 230000002123 temporal effect Effects 0.000 claims description 4
- 238000005457 optimization Methods 0.000 claims description 2
- 230000000903 blocking effect Effects 0.000 abstract description 3
- 238000005516 engineering process Methods 0.000 description 5
- 238000012545 processing Methods 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 238000010168 coupling process Methods 0.000 description 3
- 238000005859 coupling reaction Methods 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000013473 artificial intelligence Methods 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000008878 coupling Effects 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 206010039203 Road traffic accident Diseases 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004590 computer program Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000003912 environmental pollution Methods 0.000 description 1
- 230000005284 excitation Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 210000005036 nerve Anatomy 0.000 description 1
- 210000004205 output neuron Anatomy 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 238000012549 training 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 or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- 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
- G05D1/0234—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means using optical markers or beacons
- G05D1/0236—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means using optical markers or beacons in combination with a laser
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- 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 or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- 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
- G05D1/0238—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means using obstacle or wall sensors
- G05D1/024—Control of position or course in two dimensions specially adapted to land vehicles using optical position detecting means using obstacle or wall sensors in combination with a laser
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- 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
- G05D1/0257—Control of position or course in two dimensions specially adapted to land vehicles using a radar
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- 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
- G05D1/0276—Control of position or course in two dimensions specially adapted to land vehicles using signals provided by a source external to the vehicle
-
- G—PHYSICS
- G05—CONTROLLING; REGULATING
- G05D—SYSTEMS FOR CONTROLLING OR REGULATING NON-ELECTRIC VARIABLES
- G05D1/00—Control of position, course or altitude of land, water, air, or space vehicles, e.g. automatic pilot
- 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
- G05D1/0276—Control of position or course in two dimensions specially adapted to land vehicles using signals provided by a source external to the vehicle
- G05D1/0278—Control of position or course in two dimensions specially adapted to land vehicles using signals provided by a source external to the vehicle using satellite positioning signals, e.g. GPS
Abstract
The invention discloses a kind of real-time acquisition device of pilotless automobile traffic information and systems, including laser range finder, GIS geography information devices, positioning device and communication device, the laser range finder, GIS geography information device and positioning device are connect with communication device respectively, laser range finder is for acquiring range data, GIS geography information devices are for obtaining geographic position data, positioning device is used to carry out satellite positioning to vehicle, the first data of acquisition are uploaded to Cloud Server using communication device, and, when pilotless automobile passes through the traffic surveillance and control system at intersection, the second data are acquired using the traffic surveillance and control system at intersection, and it is uploaded to Cloud Server.The present invention improves the real-time of pilotless automobile acquisition traffic information data, and improves the real-time of mobile terminal or equipment acquisition pilotless automobile driving road condition data, can improve urban traffic blocking situation.
Description
Technical field
The present invention relates to technical field of transportation, more specifically, are related to a kind of pilotless automobile traffic information and obtain in real time
Take device and system.
Background technology
As the development of auto industry has also manufactured a series of problems while automobile is that people bring various convenient,
Such as traffic congestion, environmental pollution, traffic accident.The traffic national conditions in China are that motor vehicle mixes row, bicycle with non-motor vehicle
Quantity is big, and vehicles number is more, and road network carrying capacity is limited etc., and in recent years, pilotless automobile technology is increasingly ripe, such as
The Chinese patent application of Publication No. CN105788330A disclose a kind of real-time road method for early warning of automatic driving vehicle and
Device.One specific implementation mode of the method includes：Acquire the driving information of automatic driving vehicle, wherein the traveling letter
Breath includes：Destination, current time, velocity information, location information and programme path；The traveling letter is sent to Cloud Server
Breath, so that the Cloud Server determines and the relevant traffic information of the programme path according to the driving information；Described in reception
The traffic information of Cloud Server feedback；The programme path of the automatic driving vehicle is adjusted according to the traffic information；Control institute
It states automatic driving vehicle to travel according to the programme path after adjustment, realizes automatic driving vehicle and jam road is evaded.But
It is, however it remains shortcoming causes for example, the real-time that traffic information obtains is poor in intersection traffic control system
Accordingly relatively slowly, the problems such as there are the period is long, vehicle delay and the traffic capacity is limited.
Invention content
It is an object of the invention to overcome the deficiencies of the prior art and provide a kind of pilotless automobile traffic informations to obtain in real time
Device and system are taken, the traffic information data of acquisition are learnt based on neural network algorithm, and can be by the road after study
The equipment that road prediction data is issued to end side so that the equipment of end side can be by shape after history road condition data and high in the clouds study
At road prediction data be transmitted to mobile terminal device or application, not only increase pilotless automobile acquisition traffic information number
According to real-time, and improve mobile terminal or equipment and obtain the real-time that pilotless automobile drives road condition data, improve
Urban traffic blocking situation.
The purpose of the present invention is achieved through the following technical solutions：A kind of pilotless automobile traffic information obtains in real time
Device is taken, is included in the equipment of end side, Airborne Lidar examining system is provided with, the Airborne Lidar examining system includes swashing
Optar, GIS geography information device, positioning device and communication device, the laser range finder, GIS geography information device and
Positioning device is connect with communication device respectively, and laser range finder is for acquiring range data, and GIS geography information devices are for obtaining
Geographic position data, positioning device is used to carry out satellite positioning to vehicle, using the communication device by the first data of acquisition
It is uploaded to Cloud Server, also, when pilotless automobile passes through the traffic surveillance and control system at intersection, is handed over using road
Traffic surveillance and control system at prong acquires the second data, and is uploaded to Cloud Server；First receiving device is arranged in Cloud Server
On, it is handled for receiving first data and second data, and based on intelligent algorithm program module, including
Following flow：
S21 chooses multiple first data and multiple second data, while using the first data and the second data as nerve net
The sample data of network sets weights error vector as e (w), weights error and function E (w) is calculated using following formula,
Wherein, eT(w) be e (w) transposed vector；
S22 calculates the Jacobian matrix Js (w) of neural network weight；
Wherein, i is i-th of neuron, and j is j-th of neuron, the weights between i-th of neuron and j-th of neuron
For wij；
S23 calculates new weight w (k+1) using following formula,
W (k+1)=w (k)-[JT(wk)J(wk)+μI]-1JT(wk)e(wk)
Wherein, μ is additional factor, and μ ＞ 0, I are unit matrix；
S24 brings new weight w (k+1) into weights error and function calculation formula in step S21, is calculated new
Weights error and function E (w)2, compare E (w)2With original E (w)1Size, if E (w)2Numerical value reduce, then utilize
Numerical value μ divided by factor θ more than 11, then, the value of w (k+1) is updated, step S21 is then gone to；If E (w)2Number
Value does not reduce, then is multiplied by a factor θ more than 1 using numerical value μ2, w (k+1) and the S23 that gos to step are not updated, work as god
When error through network reaches preset value or convergence, stop the iterative calculation of function, exports road prediction data；Transmitting device,
Equipment for the road prediction data to be transmitted to end side；Second reception device is arranged the equipment in end side, is used for
The data of Cloud Server passback are received, and are saved it in local storage.
Further, further include update module, be arranged on Cloud Server, for obtaining first data in real time, and
For obtaining second data in real time, and using the first data and the second data as new input variable, it is based on artificial intelligence
Algoritic module carries out continuous learning, continues to optimize output road prediction data.
Further, including detection module, the first mark module, the first computing module, the second computing module, first extraction
Program module, the second extraction procedure module and wireless communication device；Detection module is arranged in the equipment in end side, is used for
The history road condition data of present road whether is preserved in detection local storage, if so, then passing through extraction procedure module
The history road condition data of present road is extracted, and utilizes the first mark module, the first label is carried out according to temporal characteristics；
First computing module, the speed mean value for calculating the history road condition data on present road；Second calculates mould
Block, the running time mean value for calculating the history road condition data on present road, and data will be calculated and be stored in local deposit
In storage device；
It is gone through in different time periods by the second extraction procedure mould for extracting present road according to the first label information
The speed mean value and/or running time mean value of history road condition data；
Wireless communication device, for road prediction data and history road condition data to be transmitted to mobile terminal device or setting
Application on the mobile terminal device.
Further, the first data include any in real time running speed, residence time, running time, mileage
Kind.
Further, the second data are included in travel speed at signal lamp, running time, residence time, signal lamp week
Phase, green time, red time any one or more of.
A kind of pilotless automobile traffic information real-time acquisition system, including any described device in technical solution as above.
The beneficial effects of the invention are as follows：
(1) the present invention is based on neural network algorithms learns the traffic information data of acquisition, and can will be after study
The road prediction data equipment that is issued to end side so that the equipment of end side can learn history road condition data and high in the clouds
The road prediction data formed afterwards is transmitted to mobile terminal device or application, not only increases pilotless automobile acquisition road conditions letter
The real-time of data is ceased, and improves the real-time of mobile terminal or equipment acquisition pilotless automobile driving road condition data,
Improve urban traffic blocking situation.
(2) present invention uses improved neural network algorithm, fast convergence rate to have calculation amount small, and precision is high, consumption money
The few feature in source is especially easy to restrain so that the real-time of information exchange is significantly improved.
(3) unmanned platform is very high to the requirement of real-time of information, and therefore, the present invention is due to using improved god
Through network algorithm, operation convergence rate is improved, real-time, predetermined speed is fast, has very big meaning.
(4) present invention is acted when the history road condition data of the equipment storage to end side is handled by label
Processing step so that when calling data, program runnability is improved, and improves data-handling efficiency, enhances and sets
Standby runnability.
(5) it is acquired present invention incorporates the real time traffic data of the equipment of end side acquisition and road traffic monitoring system
Intelligent algorithm module is deployed in cloud service by input variable of the traffic data as the neural network algorithm of high in the clouds side
Device, beyond the clouds server carry out that prediction data is issued to terminal side equipment after study processing, a large amount of distributions can be focused on
The data that formula equipment or system upload reduce the cost that analyzing processing is carried out in the equipment or system of end side, improve pair
The forecasting efficiency of traffic road congestion, the traffic trip of Effective Regulation city commuter.
Description of the drawings
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 technology description to be briefly described, it should be apparent that, the accompanying drawings in the following description is only this
Some embodiments of invention without having to pay creative labor, may be used also for those of ordinary skill in the art
With obtain other attached drawings according to these attached drawings.
Fig. 1 is the structural schematic diagram of the present invention.
Fig. 2 is the method and step flow chart of the present invention.
Specific implementation mode
Technical scheme of the present invention is described in further detail below in conjunction with the accompanying drawings, but protection scope of the present invention is not limited to
It is as described below.All features disclosed in this specification, or implicit disclosed all methods or in the process the step of, in addition to mutual
Other than the feature and/or step of repulsion, it can combine in any way.
Any feature disclosed in this specification (including any accessory claim, abstract and attached drawing), except non-specifically chatting
It states, can be replaced by other alternative features that are equivalent or have similar purpose.That is, unless specifically stated, each feature is only
It is an example in a series of equivalent or similar characteristics.
Specific embodiments of the present invention are described more fully below, it should be noted that the embodiments described herein is served only for illustrating
Illustrate, is not intended to restrict the invention.In the following description, in order to provide a thorough understanding of the present invention, a large amount of spies are elaborated
Determine details.It will be apparent, however, to one skilled in the art that：This hair need not be carried out using these specific details
It is bright.In other instances, in order to avoid obscuring the present invention, well known circuit, software or method are not specifically described.
As shown in Figure 1, a kind of real-time acquisition device of pilotless automobile traffic information, is included in the equipment of end side,
It is provided with Airborne Lidar examining system, the Airborne Lidar examining system includes laser range finder, GIS geography information device, determines
Position device and communication device, the laser range finder, GIS geography information device and positioning device are connect with communication device respectively,
Laser range finder for acquiring range data, GIS geography information devices for obtaining geographic position data, positioning device for pair
Vehicle carries out satellite positioning, and the first data of acquisition are uploaded to Cloud Server using the communication device, also, drives at nobody
When sailing automobile by traffic surveillance and control system at intersection, the traffic surveillance and control system acquisition second at intersection is utilized
Data, and it is uploaded to Cloud Server；First receiving device is arranged on Cloud Server, for receiving first data and institute
The second data are stated, and are handled based on intelligent algorithm program module, including following flow：
S21 chooses multiple first data and multiple second data, while using the first data and the second data as nerve net
The sample data of network sets weights error vector as e (w), weights error and function E (w) is calculated using following formula,
Wherein, eT(w) be e (w) transposed vector；
S22 calculates the Jacobian matrix Js (w) of neural network weight；
Wherein, i is i-th of neuron, and j is j-th of neuron, the weights between i-th of neuron and j-th of neuron
For wij；
S23 calculates new weight w (k+1) using following formula,
W (k+1)=w (k)-[JT(wk)J(wk)+μI]-1JT(wk)e(wk)
Wherein, μ is additional factor, and μ ＞ 0, I are unit matrix；
S24 brings new weight w (k+1) into weights error and function calculation formula in step S21, is calculated new
Weights error and function E (w)2, compare E (w)2With original E (w)1Size, if E (w)2Numerical value reduce, then utilize
Numerical value μ divided by factor θ more than 11, then, the value of w (k+1) is updated, step S21 is then gone to；If E (w)2Number
Value does not reduce, then is multiplied by a factor θ more than 1 using numerical value μ2, w (k+1) and the S23 that gos to step are not updated, work as god
When error through network reaches preset value or convergence, stop the iterative calculation of function, exports road prediction data；Transmitting device,
Equipment for the road prediction data to be transmitted to end side；Second reception device is arranged the equipment in end side, is used for
The data of Cloud Server passback are received, and are saved it in local storage.
Further, further include update module, be arranged on Cloud Server, for obtaining first data in real time, and
For obtaining second data in real time, and using the first data and the second data as new input variable, it is based on artificial intelligence
Algoritic module carries out continuous learning, continues to optimize output road prediction data.
Further, including detection module, the first mark module, the first computing module, the second computing module, first extraction
Program module, the second extraction procedure module and wireless communication device；Detection module is arranged in the equipment in end side, is used for
The history road condition data of present road whether is preserved in detection local storage, if so, then passing through extraction procedure module
The history road condition data of present road is extracted, and utilizes the first mark module, the first label is carried out according to temporal characteristics；
First computing module, the speed mean value for calculating the history road condition data on present road；Second calculates mould
Block, the running time mean value for calculating the history road condition data on present road, and data will be calculated and be stored in local deposit
In storage device；
It is gone through in different time periods by the second extraction procedure mould for extracting present road according to the first label information
The speed mean value and/or running time mean value of history road condition data；
Wireless communication device, for road prediction data and history road condition data to be transmitted to mobile terminal device or setting
Application on the mobile terminal device.
Further, the first data include any in real time running speed, residence time, running time, mileage
Kind.
Further, the second data are included in travel speed at signal lamp, running time, residence time, signal lamp week
Phase, green time, red time any one or more of.
A kind of pilotless automobile traffic information real-time acquisition system, including any described device in technical solution as above.
As shown in Fig. 2, a kind of real-time acquisition device of pilotless automobile traffic information and system execute such as at work
Lower step：
S1 is provided with Airborne Lidar examining system in the equipment of end side, and the Airborne Lidar examining system includes swashing
Optar, GIS geography information device, positioning device and communication device, the laser range finder, GIS geography information device and
Positioning device is connect with communication device respectively, and laser range finder is for acquiring range data, and GIS geography information devices are for obtaining
Geographic position data, positioning device is used to carry out satellite positioning to vehicle, using the communication device by the first data of acquisition
It is uploaded to Cloud Server, also, when pilotless automobile passes through the traffic surveillance and control system at intersection, is handed over using road
Traffic surveillance and control system at prong acquires the second data, and is uploaded to Cloud Server；
S2, the first data and second data described in cloud server, and based on improved neural network algorithm into
Row processing, further comprises following steps：
S21 chooses multiple first data and multiple second data, while using the first data and the second data as nerve net
The sample data of network sets weights error vector as e (w), weights error and function E (w) is calculated using following formula,
Wherein, eT(w) be e (w) transposed vector；
S22 calculates the Jacobian matrix Js (w) of neural network weight；
Wherein, i is i-th of neuron, and j is j-th of neuron, the weights between i-th of neuron and j-th of neuron
For wij；
S23 calculates new weight w (k+1) using following formula,
W (k+1)=w (k)-[JT(wk)J(wk)+μI]-1JT(wk)e(wk)
Wherein, μ is additional factor, and μ ＞ 0, I are unit matrix；
S24 brings new weight w (k+1) into weights error and function calculation formula in step S21, is calculated new
Weights error and function E (w)2, compare E (w)2With original E (w)1Size, if E (w)2Numerical value reduce, then utilize
Numerical value μ divided by factor θ more than 11, then, the value of w (k+1) is updated, step S21 is then gone to；If E (w)2Number
Value does not reduce, then is multiplied by a factor θ more than 1 using numerical value μ2, w (k+1) and the S23 that gos to step are not updated, work as god
When error through network reaches preset value or convergence, stop the iterative calculation of function, exports road prediction data；
S3, the equipment that the road prediction data is transmitted to end side by Cloud Server；
The equipment of S4, end side receive the data of Cloud Server passback, and save it in local storage.
Further, further include update step S5；
S5, Cloud Server obtain first data in real time, and Cloud Server obtains second data in real time, and by first
Data and the second data are as new input variable, based on improved neural network algorithm continuous learning described in step S2, no
Disconnected optimization output road prediction data.
Further, in step s 4, to including the following steps：
The history road conditions number that present road whether is preserved in local storage detected in the equipment of end side by S41
According to, if so, the history road condition data of present road is then extracted, and according to the first label of temporal characteristics progress；If it is not,
It is directly entered step S44；
S42, first calculates, and calculates the speed mean value of the history road condition data on present road；Second calculates, and calculates
The running time mean value of history road condition data on present road, and be stored in data are calculated in local storage；
S43, according to the first label information in step S41, extraction present road is in history road conditions number in different time periods
According to speed mean value and/or running time mean value；
S44 is based on wireless communication device, by road prediction data and history road condition data be transmitted to mobile terminal device or
Application on the mobile terminal device is set.
Further, the first data include any in real time running speed, residence time, running time, mileage
Kind.
Further, the second data are included in travel speed at signal lamp, running time, residence time, signal lamp week
Phase, green time, red time any one or more of.
The basic principle of neural network algorithm, including input layer, hidden layer (learning layer) and output layer, signal is in neuron
Between positive transmission, obtain after network output with desired output compared with, obtaining error amount, error amount backpropagation again, use
Change initial weights and threshold value, continuous iteration is until the error and minimum that are exported.In this process, neural network
Weights and threshold value are constantly adjusted by the backpropagation of error, and by after adjustment weights and threshold value preserve, for pair
The variable newly inputted is calculated, and achievees the purpose that prediction.Include the following steps：
SS1：For one group of sequential value (x, y), correspond to each input variable x, all there are one desired output y is right therewith
It answers, it needs to be determined that inputting the number of nodes of neuron in the actual operation of neural network, for example, it is set to n, output neuron
Number of nodes be set as m, the number of nodes of intermediate hidden layer is set as p；
SS2：Hidden layer output calculates, according to the connection weight and threshold between the preset input of neural network and neuron
Value b can obtain the output H of intrerneuron j,
Wherein, w is weights, and b is threshold value, and f is excitation function, and calculation formula is as follows：
SS3：Output layer output calculates, and according to the result of calculation and weights of hidden layer neuron, is calculated using following formula
The output result O of entire neural networkk：
SS4：Error calculation obtains neural network output, it is compared with reality output, calculates neural network output
With the error amount e of reality output：
E=Yk-OK
SS5：Right value update, error e carry out backpropagation, while the adjustment of weights is carried out using the variable quantity of weights,
Expression formula is as follows：
wjk=wjk+ηHjek
Wherein, η is learning rate, can be adjusted according to actual conditions；
SS6：The iteration stopping if the prediction error for reaching setting is transferred to SS2 if not obtaining and continues to change
In generation, calculates.
Wherein, to initial value and threshold value, the random number between 0~1 is generally chosen, for learning rate, in the same god
It might have different learning rates through the different phase in network, can be configured according to actual conditions, generally be may be configured as
Between 0.01~0.8.
Further, the principle of improved neural network algorithm,
If the weights between neuron i and neuron j are wij, for an input variable xk, the output of network is yk, section
The output of point i is oik, to s layers of j-th of neuron, when inputting k-th of sample, the output of node j is：
Wherein,When inputting k-th of sample for s-1 layers, the output of j-th of neuron node,It is j-th of nerve
The threshold value of member；
Define error and function Ek,
The training of neural network be ask E be minimum value when weights；
It defines s layers of neural network and shares n node, the error vector of weights is e (W), is column vector, then weights miss
Difference and function E are represented by,
Wherein, the gradient of square error and function E (w) is calculated,
Wherein,
New weight w (k+1) is calculated,
W (k+1)=w (k)-[JT(wk)J(wk)+μI]-1JT(wk)e(wk)
Due to JT(wk)J(wk) not necessarily reversible, fractional increments μ I are added, for ensureing its invertibity.In the present embodiment
Remaining technical characteristic, those skilled in the art can carry out flexibly selecting according to actual conditions and different specific to meet
Actual demand.It will be apparent, however, to one skilled in the art that：This need not be carried out using these specific details
Invention.In other instances, in order to avoid obscuring the present invention, well known algorithm, method or system etc., at this are not specifically described
The claimed technical solution of claims of invention limits within technical protection scope.
For embodiment of the method above-mentioned, for simple description, therefore it is all expressed as a series of combination of actions, still
Those skilled in the art should understand that the application is not limited by the described action sequence, because according to the application, it is a certain
A little steps can be performed in other orders or simultaneously.Secondly, it those skilled in the art should also know that, is retouched in specification
The embodiment stated belongs to preferred embodiment, necessary to involved action and unit not necessarily the application.
It will be appreciated by those of skill in the art that unit described in conjunction with the examples disclosed in the embodiments of the present disclosure and
Algorithm steps can be realized with the combination of electronic hardware or computer software and electronic hardware.These functions are actually with hard
Part or software mode execute, and depend on the specific application and design constraint of technical solution.Professional technician can be with
Distinct methods are used to realize described function each specific application, but this realization should not exceed the model of the present invention
It encloses.
Disclosed system, module and method, may be implemented in other ways.For example, device described above
Embodiment, only schematically, for example, the division of the unit, can be only a kind of division of logic function, it is practical to realize
When there may be another division manner, such as multiple units or component can be combined or can be integrated into another system, or
Some features can be ignored or not executed.Another point, shown or discussed mutual coupling or direct-coupling or communication
Connection is it may be said that by some interfaces, the INDIRECT COUPLING or communication connection of device or unit can be electrical, machinery or other
Form.
The unit that the discrete parts illustrates can be or can not also receive and be physically separated, and be shown as unit
Component can be or can not receive physical unit, you can be located at a place, or may be distributed over multiple network lists
In member.Some or all of unit therein can be selected according to the actual needs to realize the purpose of the scheme of the present embodiment.
It, can be with if the function is realized in the form of SFU software functional unit and when sold or used as an independent product
It is stored in a computer readable storage medium.Based on this understanding, technical scheme of the present invention is substantially right in other words
The part of part or the technical solution that the prior art contributes can be expressed in the form of software products, the calculating
Machine software product is stored in a storage medium, including some instructions are used so that a computer equipment (can be personal
Computer, server or network equipment etc.) it performs all or part of the steps of the method described in the various embodiments of the present invention.And
Storage medium above-mentioned includes：USB flash disk, mobile hard disk, read-only memory (Read-Only Memory, ROM), random access memory
The various media that can store program code such as device (Random Access Memory, RAM), magnetic disc or CD.
One of ordinary skill in the art will appreciate that all or part of flow in the method for realization above-described embodiment, being can
It is completed with instructing relevant hardware by computer program, the program can be stored in computer read/write memory medium
In, the program is when being executed, it may include such as the flow of the embodiment of above-mentioned each method.Wherein, the storage medium can be magnetic
Dish, CD, ROM, RAM etc..
The above is only a preferred embodiment of the present invention, it should be understood that the present invention is not limited to described herein
Form is not to be taken as excluding other embodiments, and can be used for other combinations, modifications, and environments, and can be at this
In the text contemplated scope, modifications can be made through the above teachings or related fields of technology or knowledge.And those skilled in the art institute into
Capable modifications and changes do not depart from the spirit and scope of the present invention, then all should be in the protection domain of appended claims of the present invention
It is interior.
Claims (6)
1. a kind of real-time acquisition device of pilotless automobile traffic information, which is characterized in that including：
In the equipment of end side, it is provided with Airborne Lidar examining system, the Airborne Lidar examining system includes laser ranging
Instrument, GIS geography information device, positioning device and communication device, the laser range finder, GIS geography information device and positioning dress
It sets and is connect respectively with communication device, laser range finder is for acquiring range data, and GIS geography information devices are for obtaining geographical position
Data are set, and positioning device is used to carry out satellite positioning to vehicle, the first data of acquisition is uploaded to using the communication device
Cloud Server, also, when pilotless automobile passes through the traffic surveillance and control system at intersection, at intersection
Traffic surveillance and control system acquire the second data, and be uploaded to Cloud Server；
First receiving device is arranged on Cloud Server, for receiving first data and second data, and is based on people
Work intelligent algorithm program module is handled, including following flow：
S21 chooses multiple first data and multiple second data, while using the first data and the second data as neural network
Sample data sets weights error vector as e (w), weights error and function E (w) is calculated using following formula,
Wherein, eT(w) be e (w) transposed vector；
S22 calculates the Jacobian matrix Js (w) of neural network weight；
Wherein, i is i-th of neuron, and j is j-th of neuron, and the weights between i-th of neuron and j-th of neuron are
wij；
S23 calculates new weight w (k+1) using following formula,
W (k+1)=w (k)-[JT(wk)J(wk)+μI]-1JT(wk)e(wk)
Wherein, μ is additional factor, and μ ＞ 0, I are unit matrix；
S24 brings new weight w (k+1) into weights error and function calculation formula in step S21, new power is calculated
It is worth error and function E (w)2, compare E (w)2With original E (w)1Size, if E (w)2Numerical value reduce, then utilize numerical value μ
Divided by the factor θ more than 11, then, the value of w (k+1) is updated, step S21 is then gone to；If E (w)2Numerical value do not have
There is reduction, is then multiplied by a factor θ more than 1 using numerical value μ2, w (k+1) and the S23 that gos to step are not updated, work as nerve net
When the error of network reaches preset value or convergence, stop the iterative calculation of function, exports road prediction data；
Transmitting device, the equipment for the road prediction data to be transmitted to end side；
The equipment in end side is arranged in second reception device, the data for receiving Cloud Server passback, and saves it in this
In ground storage device.
2. a kind of real-time acquisition device of pilotless automobile traffic information according to claim 1, which is characterized in that also wrap
Update module is included, is arranged on Cloud Server, is counted for obtaining first data in real time, and for obtaining described second in real time
According to, and using the first data and the second data as new input variable, continuous learning is carried out based on intelligent algorithm module, no
Disconnected optimization output road prediction data.
3. a kind of real-time acquisition device of pilotless automobile traffic information according to claim 1, which is characterized in that including
Detection module, the first mark module, the first computing module, the second computing module, the first extraction procedure module, the second extraction procedure
Module and wireless communication device；Detection module is arranged in the equipment in end side, for detect in local storage whether
The history road condition data of present road is preserved, if so, then extracting the history road conditions of present road by extraction procedure module
Data, and the first mark module is utilized, carry out the first label according to temporal characteristics；
First computing module, the speed mean value for calculating the history road condition data on present road；Second computing module is used
It is stored in local storage in the running time mean value for calculating the history road condition data on present road, and by data are calculated
In；
According to the first label information, by the second extraction procedure mould, for extracting present road on history road in different time periods
The speed mean value and/or running time mean value of condition data；
Wireless communication device is being moved for road prediction data to be transmitted to mobile terminal device or be arranged with history road condition data
Application on dynamic terminal device.
4. according to a kind of real-time acquisition device of pilotless automobile traffic information of claim 1-3 any one of them, feature
It is, the first data include any one of real time running speed, residence time, running time, mileage.
5. according to a kind of real-time acquisition device of pilotless automobile traffic information of claim 1-3 any one of them, feature
Be, the second data be included in travel speed at signal lamp, running time, the residence time, signal lamp cycle, green time,
Red time any one or more of.
6. a kind of pilotless automobile traffic information real-time acquisition system, which is characterized in that including any in claim as above
Described device.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810045749.6A CN108594799A (en) | 2018-01-17 | 2018-01-17 | The real-time acquisition device of pilotless automobile traffic information and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810045749.6A CN108594799A (en) | 2018-01-17 | 2018-01-17 | The real-time acquisition device of pilotless automobile traffic information and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN108594799A true CN108594799A (en) | 2018-09-28 |
Family
ID=63608397
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810045749.6A Withdrawn CN108594799A (en) | 2018-01-17 | 2018-01-17 | The real-time acquisition device of pilotless automobile traffic information and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108594799A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109118777A (en) * | 2018-10-12 | 2019-01-01 | 鄂尔多斯市普渡科技有限公司 | One kind being based on unpiloted road condition monitoring vehicle and monitoring method |
| CN110515106A (en) * | 2019-07-23 | 2019-11-29 | 东南大学 | A kind of multi-modal vehicle locating device of the Multi-source Information Fusion that BDS, GPS are combined and localization method |
| CN111508253A (en) * | 2019-01-31 | 2020-08-07 | 斯特拉德视觉公司 | Method for providing automatic driving service platform and server using the same |
-
2018
- 2018-01-17 CN CN201810045749.6A patent/CN108594799A/en not_active Withdrawn
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109118777A (en) * | 2018-10-12 | 2019-01-01 | 鄂尔多斯市普渡科技有限公司 | One kind being based on unpiloted road condition monitoring vehicle and monitoring method |
| CN109118777B (en) * | 2018-10-12 | 2021-06-22 | 鄂尔多斯市普渡科技有限公司 | Road condition monitoring vehicle monitoring method based on unmanned driving |
| CN111508253A (en) * | 2019-01-31 | 2020-08-07 | 斯特拉德视觉公司 | Method for providing automatic driving service platform and server using the same |
| CN110515106A (en) * | 2019-07-23 | 2019-11-29 | 东南大学 | A kind of multi-modal vehicle locating device of the Multi-source Information Fusion that BDS, GPS are combined and localization method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Li et al. | Prediction for tourism flow based on LSTM neural network | |
| Chen et al. | A short-term traffic prediction model in the vehicular cyber–physical systems | |
| Singla et al. | GPS based bus tracking system | |
| Li et al. | A traffic prediction enabled double rewarded value iteration network for route planning | |
| CN109636049B (en) | Congestion index prediction method combining road network topological structure and semantic association | |
| EP3035314A1 (en) | A traffic data fusion system and the related method for providing a traffic state for a network of roads | |
| US20160125307A1 (en) | Air quality inference using multiple data sources | |
| CN113591380B (en) | Traffic flow prediction method, medium and equipment based on graph Gaussian process | |
| CN101719315A (en) | Method for acquiring dynamic traffic information based on middleware | |
| CN111582559B (en) | Arrival time estimation method and device | |
| CN112561191A (en) | Prediction model training method, prediction method, device, apparatus, program, and medium | |
| CN108594799A (en) | The real-time acquisition device of pilotless automobile traffic information and system | |
| CN110807924A (en) | Multi-parameter fusion method and system based on full-scale full-sample real-time traffic data | |
| CN110598917B (en) | Destination prediction method, system and storage medium based on path track | |
| CN109887284A (en) | A kind of smart city traffic signalization recommended method, system and device | |
| CN108334079A (en) | Pilotless automobile method for obtaining road condition information in real time | |
| CN101706888A (en) | Method for predicting travel time | |
| Jabamony et al. | IoT Based Bus Arrival Time Prediction Using Artificial Neural Network (ANN) for Smart Public Transport System (SPTS). | |
| CN112258850A (en) | Edge side multi-sensor data fusion system of vehicle-road cooperative system | |
| Lai et al. | Understanding drivers' route choice behaviours in the urban network with machine learning models | |
| CN115471805A (en) | Point cloud processing and deep learning model training method and device and automatic driving vehicle | |
| CN108364491A (en) | Driver information based reminding method based on vehicle flowrate prediction | |
| CN108376470A (en) | Traffic road congestion prediction technique based on particle cluster algorithm | |
| Lartey et al. | XGBoost: a tree-based approach for traffic volume prediction | |
| CN109685288B (en) | Distributed traffic flow prediction method and system |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20180928 |
|
| WW01 | Invention patent application withdrawn after publication |