CN107521500B - Information acquisition method and device - Google Patents
Information acquisition method and device Download PDFInfo
- Publication number
- CN107521500B CN107521500B CN201710792535.0A CN201710792535A CN107521500B CN 107521500 B CN107521500 B CN 107521500B CN 201710792535 A CN201710792535 A CN 201710792535A CN 107521500 B CN107521500 B CN 107521500B
- Authority
- CN
- China
- Prior art keywords
- road
- feature
- sound
- vehicle
- roadway characteristic
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 84
- 230000015654 memory Effects 0.000 claims description 12
- 238000004422 calculation algorithm Methods 0.000 claims description 5
- 238000004590 computer program Methods 0.000 claims description 5
- 230000007774 longterm Effects 0.000 claims 1
- 239000010426 asphalt Substances 0.000 description 13
- 238000010586 diagram Methods 0.000 description 7
- 230000006399 behavior Effects 0.000 description 6
- 230000006870 function Effects 0.000 description 6
- 238000010801 machine learning Methods 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 4
- 230000003287 optical effect Effects 0.000 description 3
- 238000004364 calculation method Methods 0.000 description 2
- 238000004891 communication Methods 0.000 description 2
- 230000007787 long-term memory Effects 0.000 description 2
- 230000005291 magnetic effect Effects 0.000 description 2
- 230000008447 perception Effects 0.000 description 2
- 241000218691 Cupressaceae Species 0.000 description 1
- 230000015572 biosynthetic process Effects 0.000 description 1
- 238000010276 construction Methods 0.000 description 1
- 239000010639 cypress oil Substances 0.000 description 1
- 238000000151 deposition Methods 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 239000003921 oil Substances 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 230000006403 short-term memory Effects 0.000 description 1
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W40/00—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models
- B60W40/02—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models related to ambient conditions
- B60W40/06—Road conditions
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W2552/00—Input parameters relating to infrastructure
- B60W2552/05—Type of road, e.g. motorways, local streets, paved or unpaved roads
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Automation & Control Theory (AREA)
- Mathematical Physics (AREA)
- Transportation (AREA)
- Mechanical Engineering (AREA)
- Traffic Control Systems (AREA)
Abstract
This application discloses information acquisition methods and device.One specific embodiment of this method includes: the sound that generates in the driving process for acquire vehicle, the feature of the sound that collected vehicle generates in the process of moving and vehicle driving the feature of road it is associated;The sound and roadway characteristic identification model generated in the process of moving based on collected vehicle, obtain vehicle driving road feature, wherein the corresponding relationship between the feature of roadway characteristic identification model instruction sound and the feature of road.Realize using roadway characteristic identification model in the case where not needing to determine the position of vehicle, obtain vehicle driving road current pavement behavior.
Description
Technical field
This application involves vehicular fields, and in particular to environment sensing field more particularly to information acquisition method and device.
Background technique
Currently, so that vehicle is more and more intelligent, vehicle can be with intelligence as sensor is increasingly being applied in vehicle
It can ground perception running environment.During vehicle perceives running environment, the perception for the pavement behavior of road is crucial ring
One of section.Currently, the mode generallyd use are as follows: manually in advance on the electronic map to the road surface of the road in each section
Situation is labeled, and the position of vehicle is determined according to GPS, determines the section where the position of vehicle, then, from electronic map
Obtain the pavement behavior of the road in the section where the position of the vehicle of mark.
However, on the one hand, the section of GPS signal can not be got because GPS signal is interfered some, can not determine vehicle
Position where section, lead to not get the road surface shape in the section where the position of the vehicle marked in electronic map
Condition.On the other hand, in the case where the pavement behavior of the road in some sections changes, by the mark road surface of electronic map
The pavement behavior in the changed section of situation, the current pavement behavior in the changed section of pavement behavior.
Invention information
This application provides a kind of information acquisition method and devices, for solving technology existing for above-mentioned background technology part
Problem.
In a first aspect, this application provides information acquisition method, this method comprises: being generated in the driving process of acquisition vehicle
Sound, the feature of the sound that collected vehicle generates in the process of moving and vehicle driving the feature of road it is related
Connection;The sound and roadway characteristic identification model generated in the process of moving based on collected vehicle, obtain vehicle driving
The feature of road, wherein the corresponding relationship between the feature of roadway characteristic identification model instruction sound and the feature of road.
Second aspect, this application provides information acquisition device, which includes: acquisition unit, is configured to collecting vehicle
Driving process in the sound that generates, the feature and vehicle driving for the sound that collected vehicle generates in the process of moving exist
Road feature it is associated;Recognition unit, be configured to the sound generated in the process of moving based on collected vehicle and
Roadway characteristic identification model, obtain vehicle driving road feature, wherein the spy of roadway characteristic identification model instruction sound
Corresponding relationship between sign and the feature of road.
Information acquisition method and device provided by the present application, the sound that generates in the driving process by acquiring vehicle, are adopted
The feature for the sound that the vehicle collected generates in the process of moving and vehicle driving the feature of road it is associated;Based on acquisition
To the sound that generates in the process of moving of vehicle and roadway characteristic identification model, obtain vehicle driving road feature,
Wherein, the corresponding relationship between the feature of roadway characteristic identification model instruction sound and the feature of road.It realizes and utilizes road
Feature identification model in the case where not needing to determine the position of vehicle, obtain vehicle driving road current road surface shape
Condition.
Detailed description of the invention
By reading a detailed description of non-restrictive embodiments in the light of the attached drawings below, the application's is other
Feature, objects and advantages will become more apparent upon:
Fig. 1 shows the exemplary system architecture of the information acquisition method or device that can be applied to the application;
Fig. 2 shows the flow charts according to one embodiment of the information acquisition method of the application;
Fig. 3 shows the structural schematic diagram of one embodiment of the information acquisition device according to the application;
Fig. 4 shows a hardware structural diagram of the vehicle suitable for the application.
Specific embodiment
The application is described in further detail with reference to the accompanying drawings and examples.It is understood that this place is retouched
The specific embodiment stated is used only for explaining related invention, rather than the restriction to the invention.It also should be noted that in order to
Convenient for description, part relevant to related invention is illustrated only in attached drawing.
It should be noted that in the absence of conflict, the features in the embodiments and the embodiments of the present application can phase
Mutually combination.The application is described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.
Fig. 1 shows the exemplary system frame of the embodiment of the information acquisition method or device that can be applied to the application
Structure.
System architecture includes vehicle 101, network 102 and server 103.Network 102 is in vehicle 101 and server
The medium of communication link is provided between 103.Network 102 can use wireless communication link.
Vehicle 101 can use the sound that generates in the driving process of microphone acquisition vehicle 101, and vehicle 101 can be transported
Between feature of the row by the road for learning the sound that vehicle driving generates on the road of type and type of Primary Construction
Corresponding relationship machine learning model.Vehicle 101 can collect when driving according on the road of multiple and different features
Sound and road feature, the machine learning model on vehicle 101 is trained.
Sound collected on the road of multiple and different features can also be sent to server 103 by vehicle 101.Clothes
Business device 103 can run the road for learning the sound that vehicle driving generates on the road of type and type feature it
Between corresponding relationship machine learning model, server 103 can be according to the collected sound of vehicle 101 and when collecting sound
Vehicle 101 travel road feature, machine learning model is trained, by after training machine learning model send
To vehicle 101.
Referring to FIG. 2, it illustrates the processes according to one embodiment of the information acquisition method of the application.It needs to illustrate
, Overall Steps or part steps in information acquisition method provided by the embodiment of the present application can be by vehicle (such as Fig. 1
In vehicle 101) execute.Method includes the following steps:
Step 201, the sound generated in the driving process of vehicle is acquired.
In the present embodiment, the sound that acquisition vehicle generates in the process of moving can cross for collecting vehicle in current line
Cheng Zhong, the tire of vehicle of the sound source near the tire of vehicle and road contact and the sound generated.Collected vehicle is expert at
The sound generated during sailing and vehicle current driving the feature of road it is associated.
For example, the feature of sound is the waveform of the sound wave of sound, the feature of road is the type on the road surface of road, such as cypress
Oil circuit, sandstone road.When vehicle driving is on road, on the road of different road surface types when driving, sound source is in vehicle for vehicle
Tire near sound sound wave waveform be it is associated with the frictional force on road surface with tire, it is correspondingly, collected
The waveform of the sound wave of sound and vehicle driving the feature of road it is associated.
In the present embodiment, it can use and to be generated in the driving process for being mounted on the microphone acquisition vehicle of vehicle bottom
Sound.For example, can use in the driving process for the microphone acquisition vehicle that the bottom of vehicle is installed close to the position of tire
The sound of generation.
In some optional implementations of the present embodiment, the feature of road includes: the road surface types of road, road
Flatness.The sound that vehicle generates in the process of moving and vehicle driving the road surface types of road, road flatness phase
Association.
Step 202, it is based on collected sound and roadway characteristic identification model, obtains the feature of road.
In the present embodiment, it after the sound generated in the driving process for acquiring vehicle by step 201, can be based on
Collected sound and roadway characteristic identification model, obtain vehicle driving road feature i.e. vehicle current driving road
The feature on road.
For example, the feature of the sound generated in the driving process of collected vehicle includes: the sound wave of collected sound
Waveform, wave crest, the numerical value of trough of waveform etc., vehicle driving the feature of road include: the road surface types of road, road
Flatness etc..Vehicle on identical road surface types but the different road of flatness when driving, the traveling of collected vehicle
The waveform of the sound wave of the sound generated in the process, the wave crest of waveform, trough numerical value be also different.Vehicle is identical in flatness
But on the different road of road surface types when driving, the waveform of the sound wave of the sound generated in the driving process of collected vehicle,
The wave crest of waveform, the numerical value of trough are also different.
In the present embodiment, roadway characteristic identification model can indicate the sound wave for the sound that vehicle driving generates on road
Waveform, the wave crest of waveform, the features such as numerical value of trough and road the features such as road surface types, flatness corresponding relationship.
For example, including multiple items of information in roadway characteristic identification model, each item of information includes the wave of the sound wave of sound
Shape, the wave crest of waveform, trough numerical value road corresponding with the waveform of the sound wave of the sound, the wave crest of waveform, the numerical value of trough
Road surface types, flatness.
The sound that vehicle driving generates on the road of type, the spy of the road of different types can be acquired respectively in advance
At least one feature in sign is different.For example, the road of two different types can for road surface types are identical but flatness not
Same road, the road of two different types can be the road that flatness is identical but road surface types are different.Vehicle can divide
It is not travelled on identical road surface types but the different road of flatness, acquisition traveling is in identical road surface types but smooth respectively
Spend the sound generated on different roads.Vehicle can travel on identical in flatness but different road surface types road respectively,
Acquisition travels the sound generated on road identical in flatness but different road surface types respectively.It will can collect each time vehicle
The feature of road of the traveling sound generated on the road of a type and the type is added to as an item of information
In the feature identification model of road.
In the present embodiment, the sound generated in the driving process by the collected vehicle of step 201 can be extracted
Feature, it is then possible to find out the sound generated in the process of moving with collected vehicle in roadway characteristic identification model
The feature of the highest sound of the characteristic similarity of sound, it is then possible to further obtain in roadway characteristic identification model with acquisition
To the feature of the corresponding road of feature of the highest sound of characteristic similarity of sound that generates in the process of moving of vehicle, can
The corresponding road of the feature of the highest sound of the characteristic similarity of sound will be generated in the process of moving with collected vehicle
The feature on road as vehicle driving road feature, that is, vehicle current driving road feature.
In some optional implementations of the present embodiment, collected vehicle is being based on using roadway characteristic identification model
The sound generated in the process of moving judge vehicle driving road feature before, can be to roadway characteristic identification model
It is trained.Vehicle can travel on the road of different types in advance, acquire vehicle driving respectively in advance in multiple types
Road on the sound that generates.
Can based on collected vehicle driving multiple types road on the feature and multiple classes of the sound that generate
The feature of the road of type is trained roadway characteristic identification model.For example, vehicle can the cypress high in flatness respectively in advance
It is travelled on the road of low three types of gravel road of the low asphalt road of oil circuit, flatness, flatness, acquires vehicle in three types
Road on travel the sound of generation.Travelled on the road of three types using collected vehicle generation sound and three
The feature of the road of type is trained roadway characteristic identification model.
For the road of a type, the sound that can be generated on the road of the type with multi collect vehicle driving is obtained
The sound generated on the road of the type to multiple vehicle drivings, using multiple collected vehicle drivings in the road of the type
The feature of the road of sound and the type that road generates, is trained to roadway characteristic identification model.
In a training process, the road of a type of the collected vehicle driving in multiple types can use
The feature of the road of the feature and the type of the sound of upper generation is trained roadway characteristic identification model.It can generate respectively
The corresponding input vector of the sound that collected vehicle driving generates on the road of the type and the road of the type are corresponding
Roadway characteristic vector.Each of input vector component corresponding one in collected vehicle driving on the road of the type
The feature of the sound of generation, the feature of the road of the corresponding the type of each of roadway characteristic vector component.Primary
In training process, which can be input to roadway characteristic identification model, obtain prediction output vector, prediction output to
One spy of the road of the type that each of amount representation in components roadway characteristic identification model is predicted based on input vector
Sign.It can be based on the difference for the prediction output vector and roadway characteristic vector that roadway characteristic identification model exports, using under gradient
Algorithm is dropped, the model parameter of roadway characteristic identification model is adjusted, thus, complete once training to roadway characteristic identification model
Journey.
For example, the feature of collected sound includes the waveform of sound wave, the wave crest of waveform, trough of collected sound
Numerical value, vehicle in advance respectively the asphalt road of flatness 80%, the asphalt road of flatness 50%, flatness 30% gravel road three
It is travelled on the road of a type, collects the sound that vehicle travels generation on the road of three types respectively.
It is corresponding defeated that the sound that collected vehicle driving generates on the asphalt road of flatness 80% can be generated respectively
The corresponding input vector of sound that incoming vector, collected vehicle driving generate on the asphalt road of flatness 50% collects
The corresponding input vector of sound that is generated on the gravel road of the gravel road of flatness 30% of vehicle driving.Collected vehicle
Travelling the corresponding input vector of sound generated on the asphalt road of flatness 80% includes to indicate vehicle driving in flatness
The component of the waveform of the sound wave of the sound generated on 80% asphalt road, the component of the numerical value of the wave crest of waveform, trough numerical value
Component, correspondingly, the corresponding roadway characteristic vector of the asphalt road of flatness 80% includes the component for indicating asphalt road, indicates flat
The component of whole degree 80%.The corresponding input vector of the sound that collected vehicle driving generates on the asphalt road of flatness 50%
Comprising indicating the component of the waveform of the sound wave of sound that vehicle driving generates on the asphalt road of flatness 50%, the wave crest of waveform
The component of numerical value, trough numerical value component.Correspondingly, the corresponding roadway characteristic vector of the asphalt road of flatness 50% includes
It indicates the component of asphalt road, indicate the component of flatness 50%.The sound generated on the gravel road of collected flatness 30%
Corresponding input vector includes the waveform of the sound wave for the sound for indicating that vehicle driving generates on the gravel road of flatness 30%
Component, the component of numerical value of the wave crest of waveform, trough numerical value component.Correspondingly, the gravel road of flatness 30% includes table
Show the component of gravel road, indicate the component of flatness 30%.
It can utilize collected vehicle driving on the road of each of multiple types type using aforesaid way
The feature of the sound of generation and the feature of road respectively repeatedly train roadway characteristic identification model.In multiple training process
Later, roadway characteristic identification model can be with the feature corresponding relationship of the feature of instruction sound and road.
In some optional implementations of the present embodiment, roadway characteristic identification model can be LSTM (Long
Short-Term Memory, shot and long term memory network) model.LSTM model is serializing model, has input length different
Serialize feature.The input of LSTM memory unit is the hidden layer of LSTM, and the input of LSTM hidden layer is the input layer of LSTM.
The hidden layer of LSTM model, the output of memory unit are also all sequences.
In a training process, the road of a type of the collected vehicle driving in multiple types can use
The feature of the road of the feature and the type of the sound of upper generation is trained LSTM model.It can generate respectively collected
The input vector of the corresponding LSTM model of the sound that vehicle driving generates on the road of the type and the road of the type are corresponding
Roadway characteristic vector.Each of the input vector of LSTM model component corresponding one in collected vehicle driving at this
The feature of the sound generated on the road of type, one of the road of each of roadway characteristic vector representation in components the type
Feature.In a training process, which can be input to LSTM model, and using the roadway characteristic vector as
The target output vector of LSTM model.After the input vector is input to LSTM model, available prediction output vector, in advance
Survey the road of the type that each of output vector representation in components roadway characteristic identification model is predicted based on input vector
A feature.The corresponding roadway characteristic vector of road of the prediction output vector and the type that can be exported based on LSTM model
Difference the model parameter of LSTM model is adjusted using gradient descent algorithm, thus, complete the once training to LSTM model
Process.
It can utilize collected vehicle driving on the road of each of multiple types type using aforesaid way
The feature of the sound of generation and the feature of road respectively repeatedly train LSTM model.After multiple training process,
LSTM model can be with the feature corresponding relationship of the feature of instruction sound and road.
In some optional implementations of the present embodiment, collected vehicle is being based on using roadway characteristic identification model
The sound generated in the process of moving judge vehicle driving road feature when, collected vehicle can be extracted and be expert at
The feature of the sound generated during sailing, generate sound that collected vehicle generates in the process of moving it is corresponding input to
Amount a, wherein spy of the sound that the collected vehicle of each of input vector representation in components generates in the process of moving
Sign.It is then possible to which the corresponding input vector of sound that collected vehicle generates in the process of moving is input to roadway characteristic
Identification model obtains output vector, wherein each of output vector representation in components vehicle driving road, that is, vehicle work as
It is preceding traveling road a feature.To, according to the vehicle driving of each representation in components road, that is, vehicle it is current
Travel road a feature, available vehicle driving all features such as road surface types, the flatness of road.
Referring to FIG. 3, this application provides a kind of information acquisition devices as the realization to method shown in above-mentioned each figure
One embodiment, the Installation practice are corresponding with embodiment of the method shown in Fig. 2.
As shown in figure 3, information acquisition device includes: acquisition unit 301, recognition unit 302.Wherein, acquisition unit 301 is matched
Set the sound generated in the driving process for acquiring vehicle, the feature for the sound that collected vehicle generates in the process of moving
With vehicle driving the feature of road it is associated;Recognition unit 302 is configured to based on collected vehicle in driving process
The sound and roadway characteristic identification model of middle generation, obtain vehicle driving road feature, wherein roadway characteristic identify mould
Corresponding relationship between the feature of type instruction sound and the feature of road.
In some optional implementations of the present embodiment, the feature of road includes: the road surface types of road, road
Flatness.
In some optional implementations of the present embodiment, information acquisition device further include: training unit is configured to
Before the sound generated in the driving process of acquisition vehicle, acquire what vehicle driving generated on the road of multiple types respectively
Sound, wherein the road and other kinds of road of each of multiple types type have at least one different spy
Sign;The spy of the road of the feature and multiple types of the sound generated on the road of multiple types based on collected vehicle driving
Sign, is trained roadway characteristic identification model.
In some optional implementations of the present embodiment, training unit is further configured to: being generated collected
The corresponding input vector of sound generated on the road of a type of the vehicle driving in the road of multiple types, wherein defeated
The feature of one sound of each of incoming vector representation in components;Generate the corresponding roadway characteristic of road of the type to
Amount a, wherein feature of the road of each of roadway characteristic vector representation in components the type;Input vector is input to
Roadway characteristic identification model obtains prediction output vector, wherein each of prediction output vector representation in components roadway characteristic
One feature of the road of the type that identification model is predicted based on input vector;It is special based on prediction output vector and the road
The difference for levying vector, using the model parameter of gradient descent algorithm adjustment roadway characteristic identification model.
In some optional implementations of the present embodiment, recognition unit includes: model identification subelement, is configured to
The feature for extracting the sound that collected vehicle generates in the process of moving generates collected vehicle and generates in the process of moving
The corresponding input vector of sound, wherein the collected vehicle of each of input vector representation in components is in the process of moving
One feature of the sound of generation;Input vector is input to roadway characteristic identification model, obtains output vector, wherein output
Each of vector representation in components vehicle driving road a feature.
In some optional implementations of the present embodiment, roadway characteristic identification model is shot and long term memory network mould
Type.
Referring to FIG. 4, it illustrates a hardware structural diagrams of the vehicle for being suitable for the application.
As shown in figure 4, vehicle includes CPU401, memory 402, microphone 403, GPS404, microphone 403 be can be set
In the bottom of vehicle close to the position of the tire of vehicle.CPU401, memory 402, microphone 403, GPS404 pass through bus 405
It is connected with each other.It may be implemented as computer program according to the information acquisition method of the application, comprising upper in the computer program
State the instruction of operation described in embodiment.Computer program can store in memory 402.The CPU 401 of vehicle passes through
The computer program stored in memory 402 is called, the sound generated in the process of moving according to vehicle determines that vehicle driving exists
Road feature.
Present invention also provides a kind of vehicle, which can be configured with one or more processors;Memory, for depositing
One or more programs are stored up, may include in one or more programs to execute operation described in above-mentioned steps 201-202
Instruction.When one or more programs are executed by one or more processors, so that one or more processors execution is above-mentioned
It is operated described in step 201-202.
Present invention also provides a kind of computer-readable medium, which can be included in vehicle
's;It is also possible to individualism, without in supplying vehicle.Above-mentioned computer-readable medium carries one or more program,
When one or more program is executed by vehicle, so that vehicle: acquiring the sound generated in the driving process of vehicle, collect
The feature of sound that generates in the process of moving of vehicle and vehicle driving the feature of road it is associated;Based on collected
The sound and roadway characteristic identification model that vehicle generates in the process of moving, obtain vehicle driving road feature, wherein
Corresponding relationship between the feature of roadway characteristic identification model instruction sound and the feature of road.
It should be noted that computer-readable medium described herein can be computer-readable signal media or meter
Calculation machine readable storage medium storing program for executing either the two any combination.Computer readable storage medium for example may include but unlimited
In the system of electricity, magnetic, optical, electromagnetic, infrared ray or semiconductor, device or device, or any above combination.Computer can
The more specific example for reading storage medium can include but is not limited to: electrical connection, portable meter with one or more conducting wires
Calculation machine disk, hard disk, random access storage device (RAM), read-only memory (ROM), erasable programmable read only memory
(EPROM or flash memory), optical fiber, portable compact disc read-only memory (CD-ROM), light storage device, magnetic memory device or
The above-mentioned any appropriate combination of person.In this application, computer readable storage medium can be it is any include or storage program
Tangible medium, which can be commanded execution system, device or device use or in connection.And in this Shen
Please in, computer-readable signal media may include in a base band or as carrier wave a part propagate data-signal,
In carry computer-readable program code.The data-signal of this propagation can take various forms, including but not limited to
Electromagnetic signal, optical signal or above-mentioned any appropriate combination.Computer-readable signal media can also be computer-readable
Any computer-readable medium other than storage medium, the computer-readable medium can send, propagate or transmit for by
Instruction execution system, device or device use or program in connection.The journey for including on computer-readable medium
Sequence code can transmit with any suitable medium, including but not limited to: wireless, electric wire, optical cable, RF etc. are above-mentioned
Any appropriate combination.
Flow chart and block diagram in attached drawing are illustrated according to the system of the various embodiments of the application, method and computer journey
The architecture, function and operation in the cards of sequence product.In this regard, each box in flowchart or block diagram can generation
A part of one module, program segment or code of table, a part of the module, program segment or code include one or more use
The executable instruction of the logic function as defined in realizing.It should also be noted that in some implementations as replacements, being marked in box
The function of note can also occur in a different order than that indicated in the drawings.For example, two boxes succeedingly indicated are actually
It can be basically executed in parallel, they can also be executed in the opposite order sometimes, and this depends on the function involved.Also it to infuse
Meaning, the combination of each box in block diagram and or flow chart and the box in block diagram and or flow chart can be with holding
The dedicated hardware based system of functions or operations as defined in row is realized, or can use specialized hardware and computer instruction
Combination realize.
Being described in unit involved in the embodiment of the present application can be realized by way of software, can also be by hard
The mode of part is realized.Described unit also can be set in the processor, for example, can be described as: a kind of processor packet
Include acquisition unit, recognition unit.Wherein, the title of these units does not constitute the limit to the unit itself under certain conditions
It is fixed, for example, acquisition unit is also described as " unit of the sound generated in the driving process for acquiring vehicle ".
Above description is only the preferred embodiment of the application and the explanation to institute's application technology principle.Those skilled in the art
Member is it should be appreciated that invention scope involved in the application, however it is not limited to technology made of the specific combination of above-mentioned technical characteristic
Scheme, while should also cover in the case where not departing from the inventive concept, it is carried out by above-mentioned technical characteristic or its equivalent feature
Any combination is closed and other technical solutions of formation.Such as features described above and (but being not limited to) disclosed herein have it is similar
The technical characteristic of function is replaced mutually and the technical solution that is formed.
Claims (10)
1. a kind of information acquisition method, which is characterized in that the described method includes:
Acquire the sound that generates in the driving process of vehicle, the feature of the sound and vehicle driving the feature of road it is related
Connection;
Based on the sound and roadway characteristic identification model, the feature of the road is obtained, wherein the roadway characteristic identifies mould
Corresponding relationship between the feature of type instruction sound and the feature of road;
Wherein, it is based on the sound and roadway characteristic identification model, the feature for obtaining the road includes:
The feature for extracting the sound generates the corresponding input vector of the sound, wherein each of described input vector
One feature of sound described in representation in components;
The input vector is input to roadway characteristic identification model, obtains output vector, wherein is in the output vector every
One-component indicates a feature of the road.
2. the method according to claim 1, wherein the feature of road includes: the road surface types of road, road
Flatness.
3. according to the method described in claim 2, it is characterized in that, in the driving process of acquisition vehicle the sound that generates it
Before, the method also includes:
The sound that acquisition vehicle driving generates on the road of multiple types respectively, wherein each of multiple types type
Road and other kinds of road have at least one different feature;
The road of the feature and multiple types of the sound generated on the road of multiple types based on collected vehicle driving
Feature is trained roadway characteristic identification model.
4. according to the method described in claim 3, it is characterized in that, based on collected vehicle driving multiple types road
The feature of the road of the feature of the sound of upper generation and multiple types, is trained roadway characteristic identification model and includes:
It is corresponding to generate the sound generated on the road of a type of the collected vehicle driving in the road of multiple types
Input vector a, wherein feature of sound described in each of described input vector representation in components;
Generate the corresponding roadway characteristic vector of road of the type, wherein each of described roadway characteristic vector component
Indicate a feature of the road of the type;
The input vector is input to roadway characteristic identification model, obtains prediction output vector, wherein in prediction output vector
Predicted based on the input vector one of the road of the type of each representation in components roadway characteristic identification model
Feature;
Based on the difference of prediction output vector and the roadway characteristic vector, using gradient descent algorithm adjustment roadway characteristic identification
The model parameter of model.
5. method described in one of -4 according to claim 1, which is characterized in that the roadway characteristic identification model is shot and long term note
Recall network model.
6. a kind of information acquisition device, which is characterized in that described device includes:
Acquisition unit is configured to the sound generated in the driving process of acquisition vehicle, the feature and vehicle driving of the sound
Road feature it is associated;
Recognition unit is configured to obtain the feature of the road based on the sound and roadway characteristic identification model, wherein
Corresponding relationship between the feature of the roadway characteristic identification model instruction sound and the feature of road;
Wherein, recognition unit includes:
Model identifies subelement, is configured to extract the feature of the sound, generates the corresponding input vector of the sound,
In, a feature of sound described in each of described input vector representation in components;The input vector is input to road
Feature identification model, obtains output vector, wherein a spy of road described in each of described output vector representation in components
Sign.
7. device according to claim 6, which is characterized in that described device further include:
Training unit is configured to before the sound generated in the driving process of acquisition vehicle, is acquired vehicle driving respectively and is existed
The sound generated on the road of multiple types, wherein the road and other kinds of road of each of multiple types type
With at least one different feature;Spy based on the sound that collected vehicle driving generates on the road of multiple types
The feature of the road for multiple types of seeking peace, is trained roadway characteristic identification model.
8. device according to claim 7, which is characterized in that training unit is further configured to: generating collected
The corresponding input vector of sound generated on the road of a type of the vehicle driving in the road of multiple types, wherein institute
State a feature of sound described in each of input vector representation in components;The corresponding road of road for generating the type is special
Levy vector, wherein a feature of the road of type described in each of described roadway characteristic vector representation in components;It will be described
Input vector is input to roadway characteristic identification model, obtains prediction output vector, wherein each of prediction output vector point
One feature of the road for the type that amount expression roadway characteristic identification model is predicted based on the input vector;Based on pre-
The difference for surveying output vector and the roadway characteristic vector, using the model of gradient descent algorithm adjustment roadway characteristic identification model
Parameter.
9. a kind of vehicle characterized by comprising
One or more processors;
Memory, for storing one or more programs,
When one or more of programs are executed by one or more of processors, so that one or more of processors
Realize such as method as claimed in any one of claims 1 to 5.
10. a kind of computer readable storage medium, which is characterized in that be stored thereon with computer program, which is characterized in that the journey
Such as method as claimed in any one of claims 1 to 5 is realized when sequence is executed by processor.
Priority Applications (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710792535.0A CN107521500B (en) | 2017-09-05 | 2017-09-05 | Information acquisition method and device |
PCT/CN2018/098636 WO2019047645A1 (en) | 2017-09-05 | 2018-08-03 | Information acquisition method and device |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201710792535.0A CN107521500B (en) | 2017-09-05 | 2017-09-05 | Information acquisition method and device |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107521500A CN107521500A (en) | 2017-12-29 |
CN107521500B true CN107521500B (en) | 2019-11-05 |
Family
ID=60683418
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201710792535.0A Active CN107521500B (en) | 2017-09-05 | 2017-09-05 | Information acquisition method and device |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN107521500B (en) |
WO (1) | WO2019047645A1 (en) |
Families Citing this family (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107521500B (en) * | 2017-09-05 | 2019-11-05 | 百度在线网络技术(北京)有限公司 | Information acquisition method and device |
CN109785460A (en) * | 2019-01-03 | 2019-05-21 | 深圳壹账通智能科技有限公司 | Vehicle trouble recognition methods, device, computer equipment and storage medium |
US11449543B2 (en) | 2020-05-06 | 2022-09-20 | Here Global B.V. | Method, apparatus, and computer program product for vehicle localization via amplitude audio features |
US11302345B2 (en) | 2020-05-06 | 2022-04-12 | Here Global B.V. | Method, apparatus, and computer program product for vehicle localization via frequency audio features |
CN112200049B (en) * | 2020-09-30 | 2023-03-31 | 华人运通(上海)云计算科技有限公司 | Method, device and equipment for marking road surface topography data and storage medium |
CN112179299B (en) * | 2020-10-10 | 2022-10-21 | 孙树光 | Acoustic emission-based contact network smoothness detection device and method |
CN113255938A (en) * | 2021-05-07 | 2021-08-13 | 安徽创都建设集团有限公司 | Pavement base maintenance method |
CN115762565B (en) * | 2023-01-06 | 2023-05-30 | 江苏省气象服务中心 | Road noise frequency analysis-based road surface meteorological condition identification method and system |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP2343522A1 (en) * | 2008-10-30 | 2011-07-13 | Bridgestone Corporation | Method of estimating road surface condition |
ES2390302A1 (en) * | 2012-07-04 | 2012-11-08 | Universidad Politécnica de Madrid | Embarked system in vehicles and method for the detection of the state of the asphalt (Machine-translation by Google Translate, not legally binding) |
CN103003110A (en) * | 2010-05-19 | 2013-03-27 | 株式会社普利司通 | Method for estimating condition of road surface |
CN105844211A (en) * | 2015-01-29 | 2016-08-10 | 通用汽车环球科技运作有限责任公司 | System and method for classifying a road surface |
EP3173306A1 (en) * | 2015-11-27 | 2017-05-31 | Continental Automotive GmbH | Method and device for determining a type of the road on which a vehicle is driving |
Family Cites Families (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
JP4046059B2 (en) * | 2002-11-08 | 2008-02-13 | 株式会社豊田中央研究所 | Road surface condition estimation device |
KR101618551B1 (en) * | 2014-07-02 | 2016-05-09 | 엘지전자 주식회사 | Driver assistance apparatus and Vehicle including the same |
CN105667517A (en) * | 2016-01-06 | 2016-06-15 | 吴晓白 | Simple road condition flatness alarming module |
CN107521500B (en) * | 2017-09-05 | 2019-11-05 | 百度在线网络技术(北京)有限公司 | Information acquisition method and device |
-
2017
- 2017-09-05 CN CN201710792535.0A patent/CN107521500B/en active Active
-
2018
- 2018-08-03 WO PCT/CN2018/098636 patent/WO2019047645A1/en active Application Filing
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
EP2343522A1 (en) * | 2008-10-30 | 2011-07-13 | Bridgestone Corporation | Method of estimating road surface condition |
CN103003110A (en) * | 2010-05-19 | 2013-03-27 | 株式会社普利司通 | Method for estimating condition of road surface |
ES2390302A1 (en) * | 2012-07-04 | 2012-11-08 | Universidad Politécnica de Madrid | Embarked system in vehicles and method for the detection of the state of the asphalt (Machine-translation by Google Translate, not legally binding) |
CN105844211A (en) * | 2015-01-29 | 2016-08-10 | 通用汽车环球科技运作有限责任公司 | System and method for classifying a road surface |
EP3173306A1 (en) * | 2015-11-27 | 2017-05-31 | Continental Automotive GmbH | Method and device for determining a type of the road on which a vehicle is driving |
Also Published As
Publication number | Publication date |
---|---|
WO2019047645A1 (en) | 2019-03-14 |
CN107521500A (en) | 2017-12-29 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107521500B (en) | Information acquisition method and device | |
Munoz-Organero et al. | Automatic detection of traffic lights, street crossings and urban roundabouts combining outlier detection and deep learning classification techniques based on GPS traces while driving | |
US20200317200A1 (en) | Traffic obstruction detection | |
US20190295003A1 (en) | Method, apparatus, and system for in-vehicle data selection for feature detection model creation and maintenance | |
CN104737523B (en) | The situational model in mobile device is managed by assigning for the situation label of data clustering | |
CN100555355C (en) | The method and system that the passage rate of road traffic calculates and mates | |
CN110753892A (en) | Method and system for instant object tagging via cross-modality verification in autonomous vehicles | |
CN106991159B (en) | Method and apparatus for generating information | |
CN110869936A (en) | Method and system for distributed learning and adaptation in autonomous vehicles | |
CN108422949B (en) | For the information sharing method of automatic driving vehicle, device, system and equipment | |
CN110753953A (en) | Method and system for object-centric stereo vision in autonomous vehicles via cross-modality verification | |
CN109902899A (en) | Information generating method and device | |
US11710239B2 (en) | Method, apparatus, and system using a machine learning model to segment planar regions | |
EP3443482A1 (en) | Classifying entities in digital maps using discrete non-trace positioning data | |
CN109910896A (en) | Congestion in road prediction technique and device | |
CN108228887A (en) | For generating the method and apparatus of information | |
CN107257379A (en) | Method and apparatus for pushed information | |
CN109685589A (en) | Method and apparatus for pushed information | |
CN106781507A (en) | Method and apparatus for generating traffic information and test road conditions identifying system | |
CN113763700A (en) | Information processing method, information processing device, computer equipment and storage medium | |
CN107515607A (en) | Control method and device for unmanned vehicle | |
CN111372188B (en) | Method and device for determining hot spot track in area, storage medium and electronic device | |
Belenguer et al. | Vehicle identification by means of radio-frequency-identification cards and magnetic loops | |
CN110196440A (en) | It is overlapped recognition methods and device, electronic equipment, the storage medium of track | |
Oplenskedal et al. | DeepMatch: Deep matching for in-vehicle presence detection in transportation |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
GR01 | Patent grant | ||
GR01 | Patent grant |