WO2019047645A1 - 信息获取方法和装置 - Google Patents

信息获取方法和装置 Download PDF

Info

Publication number
WO2019047645A1
WO2019047645A1 PCT/CN2018/098636 CN2018098636W WO2019047645A1 WO 2019047645 A1 WO2019047645 A1 WO 2019047645A1 CN 2018098636 W CN2018098636 W CN 2018098636W WO 2019047645 A1 WO2019047645 A1 WO 2019047645A1
Authority
WO
WIPO (PCT)
Prior art keywords
road
feature
vehicle
sound
recognition model
Prior art date
Application number
PCT/CN2018/098636
Other languages
English (en)
French (fr)
Inventor
郑超
郁浩
张云飞
姜雨
闫泳杉
唐坤
Original Assignee
百度在线网络技术（北京）有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 百度在线网络技术（北京）有限公司 filed Critical 百度在线网络技术（北京）有限公司
Publication of WO2019047645A1 publication Critical patent/WO2019047645A1/zh

Links

Images

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

Definitions

  • the present application relates to the field of vehicles, and in particular to the field of environment sensing, and in particular, to an information acquisition method and apparatus.
  • the perception of the road surface condition of the road is one of the key links.
  • the commonly used method is: manually marking the road surface condition of each road section on the electronic map in advance, determining the position of the vehicle according to the GPS, determining the road section where the vehicle is located, and then obtaining the annotation from the electronic map. The road surface condition of the road where the location of the vehicle is located.
  • the road section where the position of the vehicle is located cannot be determined, and the road surface condition of the road section where the position of the vehicle marked in the electronic map is not obtained cannot be obtained.
  • the road surface condition of the road of some road sections changes, the road surface condition of the road section where the road surface condition changes by the electronic map, and the current road surface condition of the road section whose road surface condition changes.
  • the present application provides an information acquisition method and apparatus for solving the technical problems existing in the above background art.
  • the present application provides an information acquisition method, the method comprising: collecting a sound generated during a running of a vehicle, and collecting a feature of a sound generated by the collected vehicle during driving is associated with a feature of a road on which the vehicle is traveling; The feature of the road on which the vehicle is traveling is obtained based on the collected sound and road feature recognition model generated by the vehicle during traveling, wherein the road feature recognition model indicates a correspondence relationship between the feature of the sound and the feature of the road.
  • the present application provides an information acquiring apparatus, the apparatus comprising: an collecting unit configured to collect sound generated during running of the vehicle, and the collected sound of the vehicle generated during driving and the vehicle traveling The feature of the road is associated; the identification unit is configured to obtain a feature of the road on which the vehicle is traveling based on the collected sound and road feature recognition model generated by the vehicle during the driving, wherein the road feature recognition model indicates the feature of the sound and The correspondence between the characteristics of the road.
  • the information acquiring method and device provided by the present application by collecting the sound generated during the running of the vehicle, the collected features of the sound generated by the vehicle during driving are associated with the characteristics of the road on which the vehicle is traveling; based on the collected vehicle
  • the sound and road feature recognition model generated during the running is obtained as a feature of the road on which the vehicle is traveling, wherein the road feature recognition model indicates a correspondence relationship between the feature of the sound and the feature of the road. It is realized that the road feature recognition model acquires the current road surface condition of the road on which the vehicle is traveling without determining the position of the vehicle.
  • FIG. 1 illustrates an exemplary system architecture of an information acquisition method or apparatus that can be applied to the present application
  • FIG. 2 shows a flow chart of one embodiment of an information acquisition method according to the present application
  • FIG. 3 is a schematic structural diagram of an embodiment of an information acquiring apparatus according to the present application.
  • FIG. 4 shows a schematic diagram of a hardware structure of a vehicle suitable for the present application.
  • FIG. 1 illustrates an exemplary system architecture of an embodiment of an information acquisition method or apparatus that can be applied to the present application.
  • the system architecture includes a vehicle 101, a network 102, and a server 103.
  • Network 102 is used to provide a medium for communication links between vehicle 101 and server 103.
  • Network 102 can employ a wireless communication link.
  • the vehicle 101 can collect sounds generated during the running of the vehicle 101 using a microphone, and the vehicle 101 can run a machine that is initially constructed to learn the correspondence between the sounds of the roads generated on the roads of the type of vehicles and the characteristics of the roads of the type. Learning model.
  • the vehicle 101 can train the machine learning model on the vehicle 101 based on the sounds and road features acquired while traveling on a plurality of different features of the road.
  • the vehicle 101 can also transmit the sound collected on the road of a plurality of different features to the server 103.
  • the server 103 can run a machine learning model having a correspondence relationship between learning characteristics of sounds generated by the vehicle traveling on the road of the type and roads of the type, the server 103 can according to the sound collected by the vehicle 101 and the vehicle 101 when the sound is collected.
  • the characteristics of the road on which the vehicle is traveling are trained on the machine learning model, and the trained machine learning model is transmitted to the vehicle 101.
  • FIG. 2 shows a flow of one embodiment of an information acquisition method according to the present application. It should be noted that all steps or partial steps in the information acquisition method provided by the embodiments of the present application may be performed by a vehicle (for example, the vehicle 101 in FIG. 1). The method includes the following steps:
  • Step 201 Acquire a sound generated during the running of the vehicle.
  • the sound generated by the collecting vehicle during traveling may be a sound generated by the collecting vehicle in contact with the road of the vehicle of the vehicle near the tire of the vehicle during the current running.
  • the sound generated by the collected vehicle during travel is associated with the characteristics of the road on which the vehicle is currently traveling.
  • the sound is characterized by the waveform of the sound wave of the sound
  • the feature of the road is the type of the road surface of the road, such as an asphalt road or a sand road.
  • the waveform of the sound wave of the sound of the sound source near the tire of the vehicle is associated with the friction between the tire and the road surface, and accordingly, is collected.
  • the waveform of the sound wave of the sound is associated with the characteristics of the road on which the vehicle is traveling.
  • the sound generated during the running of the vehicle can be collected using a microphone installed at the bottom of the vehicle.
  • a sound generated during the running of the vehicle can be collected using a microphone installed at a position near the bottom of the vehicle near the tire.
  • the characteristics of the road include: the type of road surface of the road, and the flatness of the road.
  • the sound generated by the vehicle during traveling is associated with the type of road surface on which the vehicle is traveling, and the flatness of the road.
  • Step 202 Obtain a feature of the road based on the collected sound and road feature recognition model.
  • the feature of the road on which the vehicle is traveling that is, the feature of the road on which the vehicle is currently traveling may be obtained.
  • the characteristics of the sound generated during the running of the collected vehicle include: the waveform of the sound wave of the collected sound, the peak of the waveform, the value of the trough, etc.
  • the characteristics of the road on which the vehicle travels include: road type of the road, road The flatness and so on.
  • the waveform of the sound wave generated by the collected vehicle during the running, the peak of the waveform, and the value of the trough are also different.
  • the waveform of the sound wave generated by the collected vehicle during the running, the peak of the waveform, and the value of the trough are also different.
  • the road feature recognition model may indicate a correspondence relationship between a waveform of a sound wave of a sound generated by the vehicle on the road, a peak of the waveform, a value of the trough, and the like, and a feature such as a road surface type and a flatness of the road.
  • the road feature recognition model includes a plurality of information items, each of which includes a waveform of a sound wave of the sound, a peak of the waveform, a value of the trough, and a waveform of the sound wave of the sound, a peak of the waveform, and a value of the valley corresponding to the value of the road.
  • Road type flatness.
  • the sound generated by the vehicle traveling on the type of road may be separately collected in advance, and at least one of the features of the different types of roads is different.
  • two different types of roads may be roads of the same road type but different flatness
  • two different types of roads may be roads of the same flatness but different road types.
  • the vehicles can travel on roads of the same road type but with different flatness, respectively, and collect sounds generated on roads of the same road type but with different flatness.
  • the vehicle can be driven on roads with the same flatness but different road types, and the sounds generated on the roads with the same flatness but different road types are collected separately.
  • the sound generated each time the vehicle is traveling on one type of road and the feature of the road of this type can be added as an information item to the road feature recognition model.
  • the characteristics of the sound generated during the running of the vehicle collected by step 201 can be extracted, and then the sound generated in the road feature recognition model and the collected vehicle during the running can be found.
  • the feature of the sound having the highest feature similarity, and then the feature of the road corresponding to the feature of the sound having the highest feature similarity of the sound generated by the collected vehicle during the travel can be further obtained in the road feature recognition model, and can be
  • the feature of the road corresponding to the feature of the sound having the highest feature similarity of the sound generated by the collected vehicle is the feature of the road on which the vehicle is traveling, that is, the feature of the road on which the vehicle is currently traveling.
  • the road feature recognition model may be trained prior to utilizing the road feature recognition model to determine the characteristics of the road on which the vehicle is traveling based on the collected sounds of the vehicle during travel.
  • the vehicle can be driven on different types of roads in advance, and the sounds generated by the vehicles traveling on various types of roads are separately collected in advance.
  • the road feature recognition model can be trained based on the characteristics of the sounds generated on the plurality of types of roads on which the collected vehicles are traveling and the characteristics of the plurality of types of roads.
  • the vehicle can be driven in advance on three types of roads: a flat road with high flatness, a low-level asphalt road, and a low flat sand road, and collect sounds generated by the vehicle traveling on three types of roads.
  • the road feature recognition model is trained using the generated sound of the vehicle traveling on three types of roads and the characteristics of three types of roads.
  • the sound generated by the vehicle traveling on the road of this type can be collected multiple times, and the sound generated by the plurality of vehicles traveling on the road of the type can be obtained, and the plurality of collected vehicles are used to travel on the road of the type.
  • the characteristics of the generated sound and the characteristics of the road of this type are trained on the road feature recognition model.
  • the road feature recognition model can be trained using the characteristics of the sound generated by the collected vehicle traveling on one of the plurality of types of roads and the characteristics of the road of the type.
  • An input vector corresponding to the sound generated by the collected vehicle traveling on the road of the type and a road feature vector corresponding to the road of the type may be separately generated.
  • Each component in the input vector corresponds to a feature of the sound generated by the acquired vehicle traveling on that type of road, each component of the road feature vector corresponding to a feature of the road of that type.
  • the input vector can be input to the road feature recognition model to obtain a predicted output vector, and each component in the predicted output vector represents a feature of the road type predicted by the road feature recognition model based on the input vector.
  • the gradient descent algorithm is used to adjust the model parameters of the road feature recognition model, thereby completing a training process for the road feature recognition model.
  • the characteristics of the collected sound include the waveform of the sound wave of the collected sound, the peak of the waveform, and the value of the trough.
  • the vehicle is pre-arranged in an asphalt road with a flatness of 80%, a paved road with a flatness of 50%, and a flatness of 30%.
  • the three types of roads on the gravel road travel, collecting the sounds generated by the vehicle on three types of roads.
  • the input vector corresponding to the sound generated by the collected vehicle traveling on the asphalt road with 80% flatness can be separately generated, the input vector corresponding to the sound generated by the collected vehicle traveling on the asphalt road with a flatness of 50%, and the collected vehicle traveling.
  • the input vector corresponding to the sound generated by the collected vehicle traveling on the asphalt road of 80% of the flatness includes the component of the waveform of the sound wave representing the sound generated by the vehicle traveling on the asphalt road of 80% of the flatness, the component of the numerical value of the peak of the waveform, The component of the value of the trough, correspondingly, the road feature vector corresponding to the asphalt road of 80% of the flatness includes a component representing the asphalt road and a component representing the flatness of 80%.
  • the input vector corresponding to the sound generated by the collected vehicle traveling on the asphalt road having a flatness of 50% includes a component of a waveform of a sound wave indicating a sound generated by a vehicle traveling on an asphalt road of 50% flatness, a component of a numerical value of a peak of the waveform, The component of the value of the trough.
  • the road feature vector corresponding to the asphalt road with a flatness of 50% includes a component representing the asphalt road and a component indicating a flatness of 50%.
  • the input vector corresponding to the sound generated on the sandstone road of the collected flatness of 30% includes the component of the waveform of the sound wave representing the sound generated by the vehicle running on the sandstone road of 30% of the flatness, the component of the numerical value of the peak of the waveform, and the trough The component of the value.
  • the sandstone road with a flatness of 30% contains a component representing the gravel road and a component indicating a flatness of 30%.
  • the road feature recognition model may be trained multiple times using the features of the sound generated by the collected vehicle traveling on each of the plurality of types of roads and the characteristics of the road in the above manner. After a plurality of training processes, the road feature recognition model may indicate the feature correspondence of the sound and the feature of the road.
  • the road feature recognition model may be an LSTM (Long Short-Term Memory) model.
  • the LSTM model is a serialized model with serialization features of varying input lengths.
  • the input to the LSTM memory unit is the hidden layer of the LSTM, and the input to the LSTM hidden layer is the input layer of the LSTM.
  • the hidden layer of the LSTM model, the output of the memory unit is also a sequence.
  • the LSTM model can be trained using the features of the sound generated by the collected vehicle traveling on one of a plurality of types of roads and the characteristics of the type of road.
  • An input vector of the LSTM model corresponding to the sound generated by the collected vehicle traveling on the road of this type and a road feature vector corresponding to the road of the type may be generated separately.
  • Each component of the input vector of the LSTM model corresponds to a feature of the sound generated by the acquired vehicle traveling on that type of road, each component of the road feature vector representing a feature of the road of that type.
  • the input vector can be input to the LSTM model during a training session, and the road feature vector is used as the target output vector of the LSTM model.
  • a predicted output vector can be obtained, each component of the predicted output vector representing a feature of the type of road predicted by the road feature recognition model based on the input vector.
  • the gradient descent algorithm is used to adjust the model parameters of the LSTM model, thereby completing a training process for the LSTM model.
  • the LSTM model can be trained multiple times using the features of the sound generated by the collected vehicle traveling on each of the plurality of types of roads and the features of the road in the manner described above. After multiple training sessions, the LSTM model can indicate the characteristics of the sound and the feature correspondence of the road.
  • the collected vehicle when the road feature recognition model is used to determine the feature of the road on which the vehicle is traveling based on the collected sound generated by the vehicle during the driving, the collected vehicle may be extracted during the driving process.
  • the feature of the generated sound generates an input vector corresponding to the sound generated by the collected vehicle during travel, wherein each component of the input vector represents a feature of the collected sound generated by the vehicle during travel.
  • the input vector corresponding to the sound generated by the collected vehicle during driving may be input to the road feature recognition model to obtain an output vector, wherein each component in the output vector indicates that the vehicle on which the vehicle is traveling, that is, the vehicle currently traveling at A feature of the road. Therefore, according to one feature of the road on which the vehicle is traveling, that is, the road on which the vehicle is currently traveling, all the features such as the road type and the flatness of the road on which the vehicle is traveling can be obtained.
  • the present application provides an embodiment of an information acquiring apparatus, and the apparatus embodiment corresponds to the method embodiment shown in FIG. 2.
  • the information acquisition apparatus includes: an acquisition unit 301, and an identification unit 302.
  • the collecting unit 301 is configured to collect sounds generated during running of the vehicle, and the collected features of the generated sounds of the vehicle during driving are associated with features of the road on which the vehicle is traveling;
  • the identifying unit 302 is configured to collect based on the collected
  • the sound and road feature recognition model generated by the arriving vehicle during driving obtains the characteristics of the road on which the vehicle is traveling, wherein the road feature recognition model indicates the correspondence between the features of the sound and the features of the road.
  • the characteristics of the road include: the type of road surface of the road, and the flatness of the road.
  • the information acquiring apparatus further includes: a training unit configured to separately collect the generated by the vehicle on the plurality of types of roads before collecting the sound generated during the running of the vehicle a sound, wherein each of the plurality of types has at least one different feature from the other types of roads; based on the collected features of the sound generated by the vehicle traveling on the plurality of types of roads and the plurality of types The characteristics of the road are trained on the road feature recognition model.
  • a training unit configured to separately collect the generated by the vehicle on the plurality of types of roads before collecting the sound generated during the running of the vehicle a sound, wherein each of the plurality of types has at least one different feature from the other types of roads; based on the collected features of the sound generated by the vehicle traveling on the plurality of types of roads and the plurality of types The characteristics of the road are trained on the road feature recognition model.
  • the training unit is further configured to: generate an input vector corresponding to the sound generated by the collected vehicle traveling on one of the plurality of types of roads, wherein the input Each component in the vector represents a feature of the sound; a road feature vector corresponding to the road of the type is generated, wherein each component of the road feature vector represents a feature of the road of the type; inputting the input vector to a road feature recognition model, wherein a predicted output vector is obtained, wherein each component in the predicted output vector represents a feature of the road type predicted by the road feature recognition model based on the input vector; based on the difference between the predicted output vector and the road feature vector Value, the gradient descent algorithm is used to adjust the model parameters of the road feature recognition model.
  • the identification unit includes: a model identification subunit configured to extract features of the collected sound generated by the vehicle during the driving process, and generate the collected vehicle to generate during the driving process. a sound corresponding to the input vector, wherein each component of the input vector represents a feature of the collected sound generated by the vehicle during travel; inputting the input vector to the road feature recognition model to obtain an output vector, wherein the output vector Each of the components represents a feature of the road on which the vehicle is traveling.
  • the road feature recognition model is a long-term and short-term memory network model.
  • FIG. 4 shows a hardware structure diagram of a vehicle suitable for the present application.
  • the vehicle includes a CPU 401, a memory 402, a microphone 403, and a GPS 404, and the microphone 403 may be disposed at a position near the bottom of the vehicle that is close to the tire of the vehicle.
  • the CPU 401, the memory 402, the microphone 403, and the GPS 404 are connected to each other through a bus 405.
  • the information acquisition method according to the present application can be implemented as a computer program containing instructions of the operations described in the above embodiments.
  • the computer program can be stored in the memory 402.
  • the CPU 401 of the vehicle determines the characteristics of the road on which the vehicle is traveling based on the sound generated by the vehicle during traveling by calling a computer program stored in the memory 402.
  • the application also provides a vehicle that can be configured with one or more processors; a memory for storing one or more programs, and one or more programs can be included to perform the steps described in steps 201-202 above
  • the instructions for the operation When one or more programs are executed by one or more processors, cause one or more processors to perform the operations described in steps 201-202 above.
  • the application also provides a computer readable medium, which may be included in a vehicle, or may be separately present, not incorporated into a vehicle.
  • the computer readable medium carries one or more programs that, when executed by the vehicle, cause the vehicle to: collect sounds generated during travel of the vehicle, and collect characteristics of sounds generated by the vehicle during travel. Correlating with a feature of a road on which the vehicle is traveling; obtaining a feature of a road on which the vehicle is traveling based on the collected sound and road feature recognition model generated by the vehicle during driving, wherein the road feature recognition model indicates characteristics and roads of the sound Correspondence between the characteristics of the feature.
  • the computer readable medium described herein may be a computer readable signal medium or a computer readable storage medium or any combination of the two.
  • a computer readable storage medium may include, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination of the above. More specific examples of computer readable storage media may include, but are not limited to, electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read only memory (ROM), erasable Programmable read only memory (EPROM or flash memory), optical fiber, portable compact disk read only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination of the foregoing.
  • a computer readable storage medium may be any tangible medium that can contain or store a program, which can be used by or in connection with an instruction execution system, apparatus or device.
  • a computer readable signal medium may include a data signal that is propagated in the baseband or as part of a carrier, carrying computer readable program code. Such propagated data signals can take a variety of forms including, but not limited to, electromagnetic signals, optical signals, or any suitable combination of the foregoing.
  • the computer readable signal medium can also be any computer readable medium other than a computer readable storage medium, which can transmit, propagate, or transport a program for use by or in connection with the instruction execution system, apparatus, or device.
  • Program code embodied on a computer readable medium can be transmitted by any suitable medium, including but not limited to wireless, wire, fiber optic cable, RF, etc., or any suitable combination of the foregoing.
  • each block of the flowchart or block diagram can represent a module, a program segment, or a portion of code that includes one or more of the logic functions for implementing the specified.
  • Executable instructions can also occur in a different order than that illustrated in the drawings. For example, two successively represented blocks may in fact be executed substantially in parallel, and they may sometimes be executed in the reverse order, depending upon the functionality involved.
  • each block of the block diagrams and/or flowcharts, and combinations of blocks in the block diagrams and/or flowcharts can be implemented in a dedicated hardware-based system that performs the specified function or operation. Or it can be implemented by a combination of dedicated hardware and computer instructions.
  • the units involved in the embodiments of the present application may be implemented by software or by hardware.
  • the described unit may also be provided in the processor, for example, as a processor comprising an acquisition unit, an identification unit.
  • the names of these units do not in some cases constitute a limitation on the unit itself.
  • the acquisition unit can also be described as “a unit for collecting sounds generated during the running of the vehicle”.

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

一种信息获取方法，包括：采集车辆的行驶过程中生成的声音，采集到的车辆在行驶过程中生成的声音的特征与车辆行驶在的道路的特征相关联；基于采集到的车辆在行驶过程中生成的声音和道路特征识别模型，得到车辆行驶在的道路的特征，其中，道路特征识别模型指示声音的特征与道路的特征之间的对应关系。还公开了一种信息获取装置。通过此方法和装置实现了利用道路特征识别模型在无需确定车辆的位置的情况下，获取车辆行驶在的道路的当前的路面状况。

Description

信息获取方法和装置
本专利申请要求于2017年9月5日提交的、申请号为201710792535.0、申请人为百度在线网络技术(北京)有限公司、发明名称为“信息获取方法和装置”的中国专利申请的优先权，该申请的全文以引用的方式并入本申请中。
技术领域
本申请涉及车辆领域，具体涉及环境感知领域，尤其涉及信息获取方法和装置。
背景技术
目前，随着传感器越来越多地应用于车辆中，使得车辆越来越智能化，车辆可以智能地感知行驶环境。在车辆感知行驶环境的过程中，对于道路的路面状况的感知是关键环节之一。目前，通常采用的方式为：以人工方式预先在电子地图上对各个路段的道路的路面状况进行标注，根据GPS确定车辆的位置，确定车辆的位置所在的路段，然后，从电子地图中获取标注的车辆的位置所在的路段的道路的路面状况。
然而，一方面，在一些因GPS信号受到干扰无法获取到GPS信号的路段，无法确定车辆的位置所在的路段，导致无法获取到电子地图中标注的车辆的位置所在的路段的路面状况。另一方面，在一些路段的道路的路面状况发生变化的情况下，依靠电子地图的标注路面状况发生变化的路段的路面状况，路面状况发生变化的路段的当前的路面状况。
发明信息
本申请提供了一种信息获取方法和装置，用于解决上述背景技术部分存在的技术问题。
第一方面，本申请提供了信息获取方法，该方法包括：采集车辆的行驶过程中生成的声音，采集到的车辆在行驶过程中生成的声音的特征与车辆行驶在的道路的特征相关联；基于采集到的车辆在行驶过程中生成的声音和道路特征识别模型，得到车辆行驶在的道路的特征，其中，道路特征识别模型指示声音的特征与道路的特征之间的对应关系。
第二方面，本申请提供了信息获取装置，该装置包括：采集单元，配置用于采集车辆的行驶过程中生成的声音，采集到的车辆在行驶过程中生成的声音的特征与车辆行驶在的道路的特征相关联；识别单元，配置用于基于采集到的车辆在行驶过程中生成的声音和道路特征识别模型，得到车辆行驶在的道路的特征，其中，道路特征识别模型指示声音的特征与道路的特征之间的对应关系。
本申请提供的信息获取方法和装置，通过采集车辆的行驶过程中生成的声音，采集到的车辆在行驶过程中生成的声音的特征与车辆行驶在的道路的特征相关联；基于采集到的车辆在行驶过程中生成的声音和道路特征识别模型，得到车辆行驶在的道路的特征，其中，道路特征识别模型指示声音的特征与道路的特征之间的对应关系。实现了利用道路特征识别模型在无需确定车辆的位置的情况下，获取车辆行驶在的道路的当前的路面状况。
附图说明
通过阅读参照以下附图所作的对非限制性实施例所作的详细描述，本申请的其它特征、目的和优点将会变得更明显：
图1示出了可以应用于本申请的信息获取方法或装置的示例性系统架构；
图2示出了根据本申请的信息获取方法的一个实施例的流程图；
图3示出了根据本申请的信息获取装置的一个实施例的结构示意图；
图4示出了适用于本申请的车辆的一个硬件结构示意图。
具体实施方式
下面结合附图和实施例对本申请作进一步的详细说明。可以理解的是，此处所描述的具体实施例仅仅用于解释相关发明，而非对该发明的限定。另外还需要说明的是，为了便于描述，附图中仅示出了与有关发明相关的部分。
需要说明的是，在不冲突的情况下，本申请中的实施例及实施例中的特征可以相互组合。下面将参考附图并结合实施例来详细说明本申请。
图1示出了可以应用于本申请的信息获取方法或装置的实施例的示例性系统架构。
系统架构包括车辆101、网络102和服务器103。网络102用以在车辆101和服务器103之间提供通信链路的介质。网络102可以采用无线通信链路。
车辆101可以利用麦克风采集车辆101的行驶过程中生成的声音，车辆101可以运行由初步构建的用于学习车辆行驶在类型的道路上生成的声音和类型的道路的特征之间的对应关系的机器学习模型。车辆101可以根据在多个不同的特征的道路上行驶时采集到的声音和道路的特征，对车辆101上的机器学习模型进行训练。
车辆101也可以将在多个不同的特征的道路上采集到的声音发送至服务器103。服务器103可以运行有用于学习车辆行驶在类型的道路上生成的声音和类型的道路的特征之间的对应关系的机器学习模型，服务器103可以根据车辆101采集到的声音和采集到声音时车辆101所行驶在的道路的特征，对机器学习模型进行训练，将训练后的机器学习模型发送至车辆101。
请参考图2，其示出了根据本申请的信息获取方法的一个实施例的流程。需要说明的是，本申请实施例所提供的信息获取方法中的全部步骤或部分步骤可以由车辆(例如图1中的车辆101)执行。该方法包括以下步骤：
步骤201，采集车辆的行驶过程中生成的声音。
在本实施例中，采集车辆在行驶过程中生成的声音可以为采集车 辆在当前行驶过程中，声源在车辆的轮胎附近的车辆的轮胎与道路接触而生成的声音。采集到的车辆在行驶过程中生成的声音与车辆当前行驶在的道路的特征相关联。
例如，声音的特征为声音的声波的波形，道路的特征为道路的路面的类型，例如柏油路、沙石路。当车辆行驶在道路上时，车辆在不同的路面类型的道路上行驶时，声源在车辆的轮胎附近的声音的声波的波形是与轮胎与路面的摩擦力相关联的，相应地，采集到的声音的声波的波形与车辆行驶在的道路的特征相关联。
在本实施例中，可以利用安装在车辆底部的麦克风采集车辆的行驶过程中生成的声音。例如，可以利用在车辆的底部靠近轮胎的位置安装的麦克风采集车辆的行驶过程中生成的声音。
在本实施例的一些可选的实现方式中，道路的特征包括：道路的路面类型、道路的平整度。车辆在行驶过程中生成的声音与车辆行驶在的道路的路面类型、道路的平整度相关联。
步骤202，基于采集到的声音和道路特征识别模型，得到道路的特征。
在本实施例中，在通过步骤201采集车辆的行驶过程中生成的声音之后，可以基于采集到的声音和道路特征识别模型，得到车辆行驶在的道路的特征即车辆当前行驶在的道路的特征。
例如，采集到的车辆的行驶过程中生成的声音的特征包括：采集到的声音的声波的波形、波形的波峰、波谷的数值等，车辆行驶在的道路的特征包括：道路的路面类型、道路的平整度等。车辆在相同的路面类型但平整度不同的道路上行驶时，采集到的车辆的行驶过程中生成的声音的声波的波形、波形的波峰、波谷的数值也是不同的。车辆在平整度相同但路面类型不同的道路上行驶时，采集到的车辆的行驶过程中生成的声音的声波的波形、波形的波峰、波谷的数值也是不同的。
在本实施例中，道路特征识别模型可以指示车辆行驶在道路上生成的声音的声波的波形、波形的波峰、波谷的数值等特征与道路的路面类型、平整度等特征的对应关系。
例如，道路特征识别模型中包含多个信息项，每一个信息项包含声音的声波的波形、波形的波峰、波谷的数值和该声音的声波的波形、波形的波峰、波谷的数值对应的道路的路面类型、平整度。
可以预先分别采集车辆行驶在类型的道路上生成的声音，不同的类型的道路的特征中的至少一个特征不同。例如，两个不同的类型的道路可以为路面类型相同但平整度不同的道路，两个不同的类型的道路可以为平整度相同但路面类型不同的道路。车辆可以分别在相同的路面类型但平整度不同的道路上行驶，分别采集行驶在相同的路面类型但平整度不同的道路上生成的声音。车辆可以分别在平整度相同但路面类型不同的道路上行驶，分别采集行驶在平整度相同但路面类型不同的道路上生成的声音。可以将每一次采集到车辆行驶在一个类型的道路上生成的声音和该类型的道路的特征作为一个信息项加入到道路特征识别模型中。
在本实施例中，可以提取出通过步骤201采集到的车辆的行驶过程中生成的声音的特征，然后，可以查找出道路特征识别模型中的与采集到的车辆在行驶过程中生成的声音的特征相似度最高的声音的特征，然后，可以进一步得到道路特征识别模型中的与采集到的车辆在行驶过程中生成的声音的特征相似度最高的声音的特征对应的道路的特征，可以将与采集到的车辆在行驶过程中生成的声音的特征相似度最高的声音的特征对应的道路的特征作为车辆行驶在的道路的特征即车辆当前行驶在的道路的特征。
在本实施例的一些可选的实现方式中，在利用道路特征识别模型基于采集到的车辆在行驶过程中生成的声音判断车辆行驶在的道路的特征之前，可以对道路特征识别模型进行训练。车辆可以预先在不同的类型的道路上行驶，预先分别采集车辆行驶在多个类型的道路上生成的声音。
可以基于采集到的车辆行驶在的多个类型的道路上生成的声音的特征和多个类型的道路的特征，对道路特征识别模型进行训练。例如，车辆可以预先分别在平整度高的柏油路、平整度低的柏油路、平整度低的砂石路三个类型的道路上行驶，采集车辆在三个类型的道路上行 驶生成的声音。利用采集到的车辆在三个类型的道路上行驶生成的声音和三个类型的道路的特征，对道路特征识别模型进行训练。
针对一个类型的道路，可以多次采集车辆行驶在该类型的道路上生成的声音，得到多个车辆行驶在该类型的道路上生成的声音，利用多个采集到的车辆行驶在该类型的道路上生成的声音和该类型的道路的特征，对对道路特征识别模型进行训练。
在一次训练过程中，可以利用采集到的车辆行驶在多个类型中的一个类型的道路上生成的声音的特征和该类型的道路的特征对道路特征识别模型进行训练。可以分别生成采集到的车辆行驶在该类型的道路上生成的声音对应的输入向量和该类型的道路对应的道路特征向量。输入向量中的每一个分量对应一个在采集到的车辆行驶在该类型的道路上生成的声音的特征，道路特征向量中的每一个分量对应一个该类型的道路的特征。在一次训练过程中，可以将该输入向量输入到道路特征识别模型，得到预测输出向量，预测输出向量中的每一个分量表示道路特征识别模型基于输入向量预测出的该类型的道路的一个特征。可以基于道路特征识别模型输出的预测输出向量和道路特征向量的差值，采用梯度下降算法，调整道路特征识别模型的模型参数，从而，完成一次对道路特征识别模型的训练过程。
例如，采集到的声音的特征包含采集到的声音的声波的波形、波形的波峰、波谷的数值，车辆预先分别在平整度80％的柏油路、平整度50％的柏油路、平整度30％的砂石路三个类型的道路上行驶，分别采集到车辆在三个类型的道路上行驶生成的声音。
可以分别生成采集到的车辆行驶在平整度80％的柏油路上生成的声音对应的输入向量、采集到的车辆行驶在平整度50％的柏油路上生成的声音对应的输入向量、采集到的车辆行驶在平整度30％的砂石路的砂石路上生成的声音对应的输入向量。采集到的车辆行驶在平整度80％的柏油路上生成的声音对应的输入向量包含表示车辆行驶在平整度80％的柏油路上生成的声音的声波的波形的分量、波形的波峰的数值的分量、波谷的数值的分量，相应的，平整度80％的柏油路对应的道路特征向量包含表示柏油路的分量、表示平整度80％的分量。采集 到的车辆行驶在平整度50％的柏油路上生成的声音对应的输入向量包含表示车辆行驶在平整度50％的柏油路上生成的声音的声波的波形的分量、波形的波峰的数值的分量、波谷的数值的分量。相应的，平整度50％的柏油路对应的道路特征向量包含表示柏油路的分量、表示平整度50％的分量。采集到的平整度30％的砂石路上生成的声音对应的输入向量包含表示车辆行驶在平整度30％的砂石路上生成的声音的声波的波形的分量、波形的波峰的数值的分量、波谷的数值的分量。相应的，平整度30％的砂石路包含表示砂石路的分量、表示平整度30％的分量。
可以采用上述方式利用采集到的车辆行驶在多个类型中的每一个类型的道路上生成的声音的特征和道路的特征分别对道路特征识别模型进行多次训练。在多次训练过程之后，道路特征识别模型可以指示声音的特征和道路的特征对应关系。
在本实施例的一些可选的实现方式中，道路特征识别模型可以为LSTM(Long Short-Term Memory，长短期记忆网络)模型。LSTM模型为序列化模型，具有输入长度不一的序列化特征。LSTM记忆单元的输入是LSTM的隐藏层，LSTM隐藏层的输入是LSTM的输入层。LSTM模型的隐藏层，记忆单元的输出也都是序列。
在一次训练过程中，可以利用采集到的车辆行驶在多个类型中的一个类型的道路上生成的声音的特征和该类型的道路的特征对LSTM模型进行训练。可以分别生成采集到的车辆行驶在该类型的道路上生成的声音对应的LSTM模型的输入向量和该类型的道路对应的道路特征向量。LSTM模型的输入向量中的每一个分量对应一个在采集到的车辆行驶在该类型的道路上生成的声音的特征，道路特征向量中的每一个分量表示该类型的道路的一个特征。在一次训练过程中，可以将该输入向量输入到LSTM模型，以及将该道路特征向量作为LSTM模型的目标输出向量。将该输入向量输入到LSTM模型之后，可以得到预测输出向量，预测输出向量中的每一个分量表示道路特征识别模型基于输入向量预测出的该类型的道路的一个特征。可以基于LSTM模型输出的预测输出向量和该类型的道路对应的道路特征向量的差值， 采用梯度下降算法，调整LSTM模型的模型参数，从而，完成一次对LSTM模型的训练过程。
可以采用上述方式利用采集到的车辆行驶在多个类型中的每一个类型的道路上生成的声音的特征和道路的特征分别对LSTM模型进行多次训练。在多次训练过程之后，LSTM模型可以指示声音的特征和道路的特征对应关系。
在本实施例的一些可选的实现方式中，在利用道路特征识别模型基于采集到的车辆在行驶过程中生成的声音判断车辆行驶在的道路的特征时，可以提取采集到的车辆在行驶过程中生成的声音的特征，生成采集到的车辆在行驶过程中生成的声音对应的输入向量，其中，输入向量中的每一个分量表示采集到的车辆在行驶过程中生成的声音的一个特征。然后，可以将采集到的车辆在行驶过程中生成的声音对应的输入向量输入到道路特征识别模型，得到输出向量，其中，输出向量中的每一个分量表示车辆行驶在的道路即车辆当前行驶在的道路的一个特征。从而，根据每一个分量表示的车辆行驶在的道路即车辆当前行驶在的道路的一个特征，可以得到车辆行驶在的道路的路面类型、平整度等所有特征。
请参考图3，作为对上述各图所示方法的实现，本申请提供了一种信息获取装置的一个实施例，该装置实施例与图2所示的方法实施例相对应。
如图3所示，信息获取装置包括：采集单元301，识别单元302。其中，采集单元301配置用于采集车辆的行驶过程中生成的声音，采集到的车辆在行驶过程中生成的声音的特征与车辆行驶在的道路的特征相关联；识别单元302配置用于基于采集到的车辆在行驶过程中生成的声音和道路特征识别模型，得到车辆行驶在的道路的特征，其中，道路特征识别模型指示声音的特征与道路的特征之间的对应关系。
在本实施例的一些可选的实现方式中，道路的特征包括：道路的路面类型、道路的平整度。
在本实施例的一些可选的实现方式中，信息获取装置还包括：训练单元，配置用于在采集车辆的行驶过程中生成的声音之前，分别采 集车辆行驶在多个类型的道路上生成的声音，其中，多个类型中的每一个类型的道路与其他类型的道路具有至少一个不相同的特征；基于采集到的车辆行驶在多个类型的道路上生成的声音的特征和多个类型的道路的特征，对道路特征识别模型进行训练。
在本实施例的一些可选的实现方式中，训练单元进一步配置用于：生成采集到的车辆行驶在多个类型的道路中的一个类型的道路上生成的声音对应的输入向量，其中，输入向量中的每一个分量表示一个所述声音的特征；生成所述类型的道路对应的道路特征向量，其中，道路特征向量中的每一个分量表示该类型的道路的一个特征；将输入向量输入到道路特征识别模型，得到预测输出向量，其中，预测输出向量中的每一个分量表示道路特征识别模型基于输入向量预测出的该类型的道路的一个特征；基于预测输出向量和该道路特征向量的差值，采用梯度下降算法调整道路特征识别模型的模型参数。
在本实施例的一些可选的实现方式中，识别单元包括：模型识别子单元，配置用于提取采集到的车辆在行驶过程中生成的声音的特征，生成采集到的车辆在行驶过程中生成的声音对应的输入向量，其中，输入向量中的每一个分量表示采集到的车辆在行驶过程中生成的声音的一个特征；将输入向量输入到道路特征识别模型，得到输出向量，其中，输出向量中的每一个分量表示车辆行驶在的道路的一个特征。
在本实施例的一些可选的实现方式中，道路特征识别模型为长短期记忆网络模型。
请参考图4，其示出了适用于本申请的车辆的一个硬件结构示意图。
如图4所示，车辆包括CPU401、存储器402、麦克风403、GPS404，麦克风403可以设置在车辆的底部靠近车辆的轮胎的位置。CPU401、存储器402、麦克风403、GPS404通过总线405彼此相连。根据本申请的信息获取方法可以被实现为计算机程序，该计算机程序中包含上述实施例中描述的操作的指令。计算机程序可以存储在存储器402中。车辆的CPU 401通过调用存储器402中存储的计算机程序，根据车辆在行驶过程中生成的声音，确定车辆行驶在的道路的特征。
本申请还提供了一种车辆，该车辆可以配置有一个或多个处理器；存储器，用于存储一个或多个程序，一个或多个程序中可以包含用以执行上述步骤201-202中描述的操作的指令。当一个或多个程序被一个或多个处理器执行时，使得一个或多个处理器执行上述步骤201-202中描述的操作。
本申请还提供了一种计算机可读介质，该计算机可读介质可以是车辆中所包括的；也可以是单独存在，未装配入车辆中。上述计算机可读介质承载有一个或者多个程序，当一个或者多个程序被车辆执行时，使得车辆：采集车辆的行驶过程中生成的声音，采集到的车辆在行驶过程中生成的声音的特征与车辆行驶在的道路的特征相关联；基于采集到的车辆在行驶过程中生成的声音和道路特征识别模型，得到车辆行驶在的道路的特征，其中，道路特征识别模型指示声音的特征与道路的特征之间的对应关系。
需要说明的是，本申请所述的计算机可读介质可以是计算机可读信号介质或者计算机可读存储介质或者是上述两者的任意组合。计算机可读存储介质例如可以包括但不限于电、磁、光、电磁、红外线、或半导体的系统、装置或器件，或者任意以上的组合。计算机可读存储介质的更具体的例子可以包括但不限于：具有一个或多个导线的电连接、便携式计算机磁盘、硬盘、随机访问存储器(RAM)、只读存储器(ROM)、可擦式可编程只读存储器(EPROM或闪存)、光纤、便携式紧凑磁盘只读存储器(CD-ROM)、光存储器件、磁存储器件、或者上述的任意合适的组合。在本申请中，计算机可读存储介质可以是任何包含或存储程序的有形介质，该程序可以被指令执行系统、装置或者器件使用或者与其结合使用。而在本申请中，计算机可读的信号介质可以包括在基带中或者作为载波一部分传播的数据信号，其中承载了计算机可读的程序代码。这种传播的数据信号可以采用多种形式，包括但不限于电磁信号、光信号或上述的任意合适的组合。计算机可读的信号介质还可以是计算机可读存储介质以外的任何计算机可读介质，该计算机可读介质可以发送、传播或者传输用于由指令执行系统、装置或者器件使用或者与其结合使用的程序。计算机可读介质 上包含的程序代码可以用任何适当的介质传输，包括但不限于：无线、电线、光缆、RF等等，或者上述的任意合适的组合。
附图中的流程图和框图，图示了按照本申请各种实施例的系统、方法和计算机程序产品的可能实现的体系架构、功能和操作。在这点上，流程图或框图中的每个方框可以代表一个模块、程序段、或代码的一部分，该模块、程序段、或代码的一部分包含一个或多个用于实现规定的逻辑功能的可执行指令。也应当注意，在有些作为替换的实现中，方框中所标注的功能也可以以不同于附图中所标注的顺序发生。例如，两个接连地表示的方框实际上可以基本并行地执行，它们有时也可以按相反的顺序执行，这依所涉及的功能而定。也要注意的是，框图和/或流程图中的每个方框、以及框图和/或流程图中的方框的组合，可以用执行规定的功能或操作的专用的基于硬件的系统来实现，或者可以用专用硬件与计算机指令的组合来实现。
描述于本申请实施例中所涉及到的单元可以通过软件的方式实现，也可以通过硬件的方式来实现。所描述的单元也可以设置在处理器中，例如，可以描述为：一种处理器包括采集单元，识别单元。其中，这些单元的名称在某种情况下并不构成对该单元本身的限定，例如，采集单元还可以被描述为“用于采集车辆的行驶过程中生成的声音的单元”。
以上描述仅为本申请的较佳实施例以及对所运用技术原理的说明。本领域技术人员应当理解，本申请中所涉及的发明范围，并不限于上述技术特征的特定组合而成的技术方案，同时也应涵盖在不脱离所述发明构思的情况下，由上述技术特征或其等同特征进行任意组合合而形成的其它技术方案。例如上述特征与本申请中公开的(但不限于)具有类似功能的技术特征进行互相替换而形成的技术方案。

Claims (12)

  1. 一种信息获取方法，其特征在于，所述方法包括：
    采集车辆的行驶过程中生成的声音，所述声音的特征与车辆行驶在的道路的特征相关联；
    基于所述声音和道路特征识别模型，得到所述道路的特征，其中，所述道路特征识别模型指示声音的特征与道路的特征之间的对应关系。
  2. 根据权利要求1所述的方法，其特征在于，道路的特征包括：道路的路面类型、道路的平整度。
  3. 根据权利要求2所述的方法，其特征在于，在采集车辆的行驶过程中生成的声音之前，所述方法还包括：
    分别采集车辆行驶在多个类型的道路上生成的声音，其中，多个类型中的每一个类型的道路与其他类型的道路具有至少一个不相同的特征；
    基于采集到的车辆行驶在多个类型的道路上生成的声音的特征和多个类型的道路的特征，对道路特征识别模型进行训练。
  4. 根据权利要求3所述的方法，其特征在于，基于采集到的车辆行驶在多个类型的道路上生成的声音的特征和多个类型的道路的特征，对道路特征识别模型进行训练包括：
    生成采集到的车辆行驶在多个类型的道路中的一个类型的道路上生成的声音对应的输入向量，其中，所述输入向量中的每一个分量表示所述声音的一个特征；
    生成所述类型的道路对应的道路特征向量，其中，所述道路特征向量中的每一个分量表示所述类型的道路的一个特征；
    将所述输入向量输入到道路特征识别模型，得到预测输出向量，其中，预测输出向量中的每一个分量表示道路特征识别模型基于所述 输入向量预测出的所述类型的道路的一个特征；
    基于预测输出向量和所述道路特征向量的差值，采用梯度下降算法调整道路特征识别模型的模型参数。
  5. 根据权利要求4所述的方法，其特征在于，基于所述声音和道路特征识别模型，得到所述道路的特征包括：
    提取所述声音的特征，生成所述声音对应的输入向量，其中，所述输入向量中的每一个分量表示所述声音的一个特征；
    将所述输入向量输入到道路特征识别模型，得到输出向量，其中，所述输出向量中的每一个分量表示所述道路的一个特征。
  6. 根据权利要求1-5之一所述的方法，其特征在于，所述道路特征识别模型为长短期记忆网络模型。
  7. 一种信息获取装置，其特征在于，所述装置包括：
    采集单元，配置用于采集车辆的行驶过程中生成的声音，所述声音的特征与车辆行驶在的道路的特征相关联；
    识别单元，配置用于基于所述声音和道路特征识别模型，得到所述道路的特征，其中，所述道路特征识别模型指示声音的特征与道路的特征之间的对应关系。
  8. 根据权利要求7所述的装置，其特征在于，所述装置还包括：
    训练单元，配置用于在采集车辆的行驶过程中生成的声音之前，分别采集车辆行驶在多个类型的道路上生成的声音，其中，多个类型中的每一个类型的道路与其他类型的道路具有至少一个不相同的特征；基于采集到的车辆行驶在多个类型的道路上生成的声音的特征和多个类型的道路的特征，对道路特征识别模型进行训练。
  9. 根据权利要求8所述的装置，其特征在于，训练单元进一步配置用于：生成采集到的车辆行驶在多个类型的道路中的一个类型的道 路上生成的声音对应的输入向量，其中，所述输入向量中的每一个分量表示所述声音的一个特征；生成所述类型的道路对应的道路特征向量，其中，所述道路特征向量中的每一个分量表示所述类型的道路的一个特征；将所述输入向量输入到道路特征识别模型，得到预测输出向量，其中，预测输出向量中的每一个分量表示道路特征识别模型基于所述输入向量预测出的所述类型的道路的一个特征；基于预测输出向量和所述道路特征向量的差值，采用梯度下降算法调整道路特征识别模型的模型参数。
  10. 根据权利要求9所述的装置，其特征在于，识别单元包括：
    模型识别子单元，配置用于提取所述声音的特征，生成所述声音对应的输入向量，其中，所述输入向量中的每一个分量表示所述声音的一个特征；将所述输入向量输入到道路特征识别模型，得到输出向量，其中，所述输出向量中的每一个分量表示所述道路的一个特征。
  11. 一种车辆，其特征在于，包括：
    一个或多个处理器；
    存储器，用于存储一个或多个程序，
    当所述一个或多个程序被所述一个或多个处理器执行时，使得所述一个或多个处理器实现如权利要求1-6中任一所述的方法。
  12. 一种计算机可读存储介质，其特征在于，其上存储有计算机程序，其特征在于，该程序被处理器执行时实现如权利要求1-6中任一所述的方法。
PCT/CN2018/098636 2017-09-05 2018-08-03 信息获取方法和装置 WO2019047645A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201710792535.0 2017-09-05
CN201710792535.0A CN107521500B (zh) 2017-09-05 2017-09-05 信息获取方法和装置

Publications (1)

Publication Number Publication Date
WO2019047645A1 true WO2019047645A1 (zh) 2019-03-14

Family

ID=60683418

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/098636 WO2019047645A1 (zh) 2017-09-05 2018-08-03 信息获取方法和装置

Country Status (2)

Country Link
CN (1) CN107521500B (zh)
WO (1) WO2019047645A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
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
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

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107521500B (zh) * 2017-09-05 2019-11-05 百度在线网络技术（北京）有限公司 信息获取方法和装置
CN109785460A (zh) * 2019-01-03 2019-05-21 深圳壹账通智能科技有限公司 车辆故障识别方法、装置、计算机设备和存储介质
CN112200049B (zh) * 2020-09-30 2023-03-31 华人运通(上海)云计算科技有限公司 对路面地形进行数据标注的方法、装置、设备及存储介质
CN112179299B (zh) * 2020-10-10 2022-10-21 孙树光 一种基于声发射的接触网平顺性检测装置及方法
CN113255938A (zh) * 2021-05-07 2021-08-13 安徽创都建设集团有限公司 一种路面基层养护方法
CN115762565B (zh) * 2023-01-06 2023-05-30 江苏省气象服务中心 一种基于路噪音频分析的路面气象状况识别方法及系统

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2004168286A (ja) * 2002-11-08 2004-06-17 Toyota Central Res & Dev Lab Inc 路面状態推定装置、路面摩擦状態推定装置、路面状態物理量演算装置、及び路面状態報知装置
CN103003110A (zh) * 2010-05-19 2013-03-27 株式会社普利司通 路面状况估计方法
CN105270263A (zh) * 2014-07-02 2016-01-27 Lg电子株式会社 车辆驾驶辅助装置及车辆
CN105667517A (zh) * 2016-01-06 2016-06-15 吴晓白 一种简易路况平整度告警模块
CN105844211A (zh) * 2015-01-29 2016-08-10 通用汽车环球科技运作有限责任公司 用于分类路面的系统和方法
CN107521500A (zh) * 2017-09-05 2017-12-29 百度在线网络技术（北京）有限公司 信息获取方法和装置

Family Cites Families (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP2343522B1 (en) * 2008-10-30 2018-04-11 Bridgestone Corporation Method of estimating road surface condition
ES2390302B2 (es) * 2012-07-04 2013-02-25 Universidad Politécnica de Madrid Sistema embarcado en vehículos y método para la detección del estado del asfalto
EP3173306B1 (en) * 2015-11-27 2019-10-30 Continental Automotive GmbH Method and device for determining a type of the road on which a vehicle is driving

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2004168286A (ja) * 2002-11-08 2004-06-17 Toyota Central Res & Dev Lab Inc 路面状態推定装置、路面摩擦状態推定装置、路面状態物理量演算装置、及び路面状態報知装置
CN103003110A (zh) * 2010-05-19 2013-03-27 株式会社普利司通 路面状况估计方法
CN105270263A (zh) * 2014-07-02 2016-01-27 Lg电子株式会社 车辆驾驶辅助装置及车辆
CN105844211A (zh) * 2015-01-29 2016-08-10 通用汽车环球科技运作有限责任公司 用于分类路面的系统和方法
CN105667517A (zh) * 2016-01-06 2016-06-15 吴晓白 一种简易路况平整度告警模块
CN107521500A (zh) * 2017-09-05 2017-12-29 百度在线网络技术（北京）有限公司 信息获取方法和装置

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
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
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

Also Published As

Publication number Publication date
CN107521500A (zh) 2017-12-29
CN107521500B (zh) 2019-11-05

Similar Documents

Publication Publication Date Title
WO2019047645A1 (zh) 信息获取方法和装置
CN111626208B (zh) 用于检测小目标的方法和装置
CN109492656B (zh) 用于输出信息的方法和装置
CN108831505B (zh) 用于识别应用的使用场景的方法和装置
US9020876B2 (en) On-demand suggestion for vehicle driving
WO2019047656A1 (zh) 用于控制无人驾驶车辆的方法和装置
CN107452369B (zh) 语音合成模型生成方法和装置
CN109121022B (zh) 用于标记视频片段的方法及装置
JP2018536236A (ja) 低解像度リモートセンシング画像を用いた移動車両の検出及び分析
CN110689804B (zh) 用于输出信息的方法和装置
CN110119725B (zh) 用于检测信号灯的方法及装置
CN114970705B (zh) 基于多传感数据的行驶状态分析方法、装置、设备及介质
CN107257379A (zh) 用于推送信息的方法和装置
WO2018105122A1 (ja) 教師データ候補抽出プログラム、教師データ候補抽出装置、及び教師データ候補抽出方法
CN116129387A (zh) 用于识别分歧路口的方法和装置
CN106095915A (zh) 用户身份的处理方法及装置
CN109146333A (zh) 导航算法评估方法和装置
CN110135517B (zh) 用于获取车辆相似度的方法及装置
CN109752018A (zh) 信息获取方法及装置
CN112598897B (zh) 交通参与者行为检测方法、装置、电子设备和介质
CN111310858B (zh) 用于生成信息的方法和装置
CN110542426B (zh) 小路识别方法、设备及可读介质
CN110119722B (zh) 用于生成信息的方法和装置
CN107084728A (zh) 用于检测数字地图的方法和装置
CN107490383B (zh) 导航中路径健康信息的服务提供方法、服务器及可读介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 18853118

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205A DATED 11/08/2020)

122 Ep: pct application non-entry in european phase

Ref document number: 18853118

Country of ref document: EP

Kind code of ref document: A1