WO2024087918A1 - 基于uwb的语音识别汽车定位方法及*** - Google Patents

基于uwb的语音识别汽车定位方法及*** Download PDF

Info

Publication number
WO2024087918A1
WO2024087918A1 PCT/CN2023/118247 CN2023118247W WO2024087918A1 WO 2024087918 A1 WO2024087918 A1 WO 2024087918A1 CN 2023118247 W CN2023118247 W CN 2023118247W WO 2024087918 A1 WO2024087918 A1 WO 2024087918A1
Authority
WO
WIPO (PCT)
Prior art keywords
car
uwb
voice
positioning
signal
Prior art date
Application number
PCT/CN2023/118247
Other languages
English (en)
French (fr)
Inventor
卜子云
Original Assignee
歌尔微电子股份有限公司
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 歌尔微电子股份有限公司 filed Critical 歌尔微电子股份有限公司
Publication of WO2024087918A1 publication Critical patent/WO2024087918A1/zh

Links

Classifications

    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L15/00—Speech recognition
    • G10L15/22—Procedures used during a speech recognition process, e.g. man-machine dialogue
    • G—PHYSICS
    • G10—MUSICAL INSTRUMENTS; ACOUSTICS
    • G10L—SPEECH ANALYSIS TECHNIQUES OR SPEECH SYNTHESIS; SPEECH RECOGNITION; SPEECH OR VOICE PROCESSING TECHNIQUES; SPEECH OR AUDIO CODING OR DECODING
    • G10L21/00—Speech or voice signal processing techniques to produce another audible or non-audible signal, e.g. visual or tactile, in order to modify its quality or its intelligibility
    • G10L21/02—Speech enhancement, e.g. noise reduction or echo cancellation
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W64/00—Locating users or terminals or network equipment for network management purposes, e.g. mobility management
    • Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
    • Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
    • Y02D—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN INFORMATION AND COMMUNICATION TECHNOLOGIES [ICT], I.E. INFORMATION AND COMMUNICATION TECHNOLOGIES AIMING AT THE REDUCTION OF THEIR OWN ENERGY USE
    • Y02D30/00—Reducing energy consumption in communication networks
    • Y02D30/70—Reducing energy consumption in communication networks in wireless communication networks

Definitions

  • the present application relates to the field of automobile application technology, and more specifically, to a UWB-based voice recognition automobile positioning method and system.
  • UWB Ultra Wide Band
  • UWB smart car keys which realize real-time high-precision positioning in combination with the UWB system inside the car, and control the car to perform specific functions based on the real-time distance between the car key and the car.
  • the in-car UWB positioning system is usually composed of 3-5 UWB positioning modules (one master node + multiple anchor points), and most of them communicate with the vehicle system through the CAN (Controller Area Network) and LIN (Local Interconnect Network) buses.
  • CAN Controller Area Network
  • LIN Local Interconnect Network
  • the purpose of this application is to provide a UWB-based voice recognition vehicle positioning method and system to solve the existing problems such as the accuracy of off-vehicle voice recognition.
  • the UWB-based voice recognition vehicle positioning method provided in this application includes:
  • the pulse signal is received by the UWB positioning module located at various positions in the car, and a timestamp of the received pulse signal is marked;
  • the DSP performs beamforming processing on the voice signal picked up by the voice pickup module to obtain an enhanced voice signal
  • the car is controlled to complete the corresponding instructions of the enhanced voice signal.
  • a preferred solution is that before sending a pulse signal to the location of the car through the car key, it also includes:
  • the car key is matched with the car by Bluetooth, and the car receives the Bluetooth signal of the car key at regular intervals;
  • Each UWB module in the car receives the pulse signal sent by the car key.
  • a preferred solution is that before processing the pulse signal and the corresponding timestamp to determine the relative position between the user carrying the car key and the car, it also includes:
  • the pulse signals received by the UWB positioning modules at each location and the corresponding timestamps are transmitted to the micro control unit via the bus;
  • the pulse signal and the corresponding time stamp are processed by the micro control unit to determine the relative position between the user carrying the car key and the car.
  • a preferred solution is to use TOA, TDOA or AOA to process the pulse signal and the corresponding time stamp to determine the relative position of the user carrying the car key and the car.
  • a preferred solution is that before the DSP performs beamforming processing on the voice signal picked up by the voice pickup module corresponding to the relative position of the user and the car to obtain the enhanced voice signal, it includes:
  • the voice signals acquired by the voice pickup modules located at various locations in the car are transmitted to the microcontroller Unit, wherein the voice pickup module at each position is set correspondingly to the UWB positioning module at each position;
  • the micro control unit performs beamforming processing and signal shielding processing on the voices acquired by the voice pickup modules at various positions.
  • micro control unit performs voice signal shielding processing on the voice recognition module that does not correspond to the user and the car.
  • the present application also provides a UWB-based voice recognition vehicle positioning system, comprising: a UWB positioning module, a voice pickup module corresponding to the UWB positioning module, a micro control unit and a vehicle computer, wherein:
  • a UWB positioning module used to obtain positioning pulse signals between the user carrying the car key and the car, and transmit the positioning pulse signals to the micro control unit;
  • a voice pickup module used to obtain a user's voice signal and transmit the voice signal to the micro control unit;
  • a micro control unit used for performing data processing on the positioning pulse signal and the voice signal, and sending the data processing result to the vehicle computer;
  • the car computer is used to send control instructions to the car based on the data processing results.
  • a preferred solution is to further include an audio bus, wherein:
  • the positioning pulse signal and the voice signal are transmitted to the micro control unit through the audio bus respectively.
  • a preferred solution is that the number of UWB positioning modules is at least 2, and the number of voice pickup modules is at least 2.
  • the UWB positioning modules and the voice pickup modules are electrically connected to each other via a connector.
  • the UWB-based voice recognition car positioning method and system provided in this application, through the combination of the UWB positioning module and the voice pickup module, accurately confirms the user's position relative to the car, realizes beamforming noise reduction, and accurately receives user commands, so as to improve the intelligence level of off-vehicle voice recognition, reduce the probability of voice command recognition errors, and efficiently complete the interaction between the car and the user.
  • FIG1 is a flow chart of a method for locating a vehicle using voice recognition based on UWB according to an embodiment of the present application
  • FIG2 is a schematic diagram of a UWB-based voice recognition vehicle positioning process according to an embodiment of the present application.
  • FIG3 is a schematic diagram of the principle of voice recognition vehicle positioning based on UWB according to an embodiment of the present application
  • FIG4 is a schematic diagram of the logical structure of a UWB-based voice recognition vehicle positioning system according to an embodiment of the present application.
  • the present application provides a UWB-based voice recognition vehicle positioning method and system.
  • FIG. 1 shows According to the UWB-based voice recognition car positioning method process of the embodiment of the present application.
  • the UWB-based voice recognition vehicle positioning method includes:
  • S120 receiving the pulse signal through the UWB positioning modules located at various positions in the car, and marking the timestamp of receiving the pulse signal;
  • S130 Processing the pulse signal and the corresponding timestamp to determine the relative position between the user carrying the car key and the car;
  • S140 Determine a voice pickup module corresponding to the relative position according to the relative position
  • S150 performing beamforming processing on the voice signal picked up by the voice pickup module through the DSP to obtain an enhanced voice signal
  • S160 According to the enhanced voice signal, controlling the vehicle to complete the instruction corresponding to the enhanced voice signal.
  • the UWB positioning module and the voice pickup module are combined with each other to achieve accurate positioning of the user's relative position to the vehicle, realize beamforming noise reduction, and receive the interactor's instructions more accurately.
  • step S110 before sending a pulse signal to the location of the car through the car key, it also includes:
  • the car key is matched with the car by Bluetooth, and the car receives the Bluetooth signal of the car key at regular intervals;
  • Each UWB module in the car receives the pulse signal sent by the car key.
  • the car key used by the user has a BLE function and can broadcast signals at a regular interval.
  • the car receives the car key BLE signal and determines whether it enters the working range of the UWB positioning module based on the RSSI signal strength.
  • the method Before processing the pulse signal and the corresponding timestamp to determine the relative position between the user carrying the car key and the car, the method further includes:
  • the pulse signals received by the UWB positioning modules at each location and the corresponding timestamps are transmitted to the micro control unit via the bus;
  • the received pulse signal and the corresponding time stamp are processed by the microcontroller unit to determine the relative position of the user and the car.
  • the relative position between the car owner and the car is determined by UWB communication between the car key and the UWB positioning module on the car.
  • the car key with UWB+BLE function is used as an anchor point.
  • BLE Bluetooth Low Energy
  • the car key automatically triggers the UWB transmission signal, and multiple UWB-based positioning modules on the car body receive the UWB signal emitted by the car key, thereby determining the relative position between the user and the car.
  • TOA Time of Arrival
  • TDOA Time Difference of Arrival
  • AOA Angle of Arrival
  • TOA Time of Arrival
  • TDOA Time Difference of Arrival
  • AOA Angle of Arrival
  • TDOA is a positioning method that infers the relative position of the object to be located relative to each reference base station (hereinafter referred to as Anchor or Anc) by solving a set of nonlinear hyperbolic equations based on the distance difference between each reference base station (hereinafter referred to as Anchor or Anc) and the object to be located (hereinafter referred to as Tag). Since the flight speed of electromagnetic waves is known and constant, the flight distance and flight time can be converted to each other, and if the transmission time is the same (or the interval is known), the flight time can be further converted to the signal arrival time. Therefore, it is only necessary to measure the difference in arrival time of the signal sent by the Tag to each Anchor to obtain the corresponding distance difference.
  • the method before beamforming processing is performed on the voice signal picked up by the voice pickup module corresponding to the relative position by the DSP to obtain the enhanced voice signal, the method includes:
  • the voice signals acquired by the voice pickup modules at various positions in the car are transmitted to the micro control unit, wherein the voice pickup modules at various positions are arranged correspondingly to the UWB positioning modules at various positions;
  • the micro control unit performs beamforming processing and signal shielding processing on the voices acquired by the voice pickup modules at various positions.
  • DSP digital signal processing
  • speech coding includes speech coding, speech synthesis, speech recognition, speech enhancement, voice mail, voice storage, etc.
  • speech enhancement technology is adopted to enhance the speech signal through DSP to obtain accurate voice commands.
  • the microcontroller unit performs voice signal shielding processing on the voice recognition module that does not correspond to the user and the car.
  • the car key used by the user has a BLE function and can broadcast signals at regular intervals.
  • the car receives the car key BLE signal and determines whether it enters the UWB working range based on the RSSI signal strength.
  • the car key When the user enters the UWB working range, the car key sends a pulse signal at a fixed time. After receiving the pulse signal from the car key, the six modules at different positions on the car will add a timestamp of the time point of receiving the pulse. The timestamps of the pulse signals received by the modules at different positions will be different.
  • the relevant data information of modules 1-6# (six UWB positioning modules and six voice pickup modules are integrated into six positioning voice modules, that is, one module includes a UWB positioning module and a voice pickup module) is transmitted to the main node module A (micro control unit) through the bus.
  • Module A realizes the positioning between users and people based on the TDOA algorithm through the received timestamp and determines the actual location of the user (as shown in Figure 3, area 1-6).
  • the car key will intermittently send out pulse signals for module A (micro control unit) to update the positioning information in real time.
  • module A micro control unit
  • modules 1-6 pick up the voice
  • the voice signals acquired from modules 1-6 are processed through the bus (A2B, etc.).
  • the user is in area 6, and module A (micro control unit) will use DSP to process the voice acquired from modules 1 and 2 using a beamforming algorithm, and enhance the voice signals in this area, and shield the voice signals acquired in the remaining areas 1-5.
  • the car will not execute the commands, but only the commands issued by the user.
  • the voice signal is transmitted to the vehicle computer, and the vehicle computer executes specific instructions according to the specific content of the voice signal.
  • the present application also provides a UWB-based voice recognition vehicle positioning system.
  • FIG4 shows the logical structure of the UWB-based voice recognition vehicle positioning system according to an embodiment of the present application.
  • the UWB-based voice recognition vehicle positioning system 400 includes: a UWB positioning module 410, a voice pickup module 420 corresponding to the UWB positioning module 410, a micro control unit 430, and a vehicle computer 440, wherein:
  • the UWB positioning module 410 is used to obtain the positioning pulse signal between the user carrying the car key and the car, and transmit the positioning pulse signal to the micro control unit;
  • the voice pickup module 420 is used to obtain the user's voice signal and transmit the voice signal to the micro control unit;
  • the micro control unit 430 is used to perform data processing on the positioning pulse signal and the voice signal, and send the data processing result to the vehicle computer;
  • the vehicle computer 440 is used to send control instructions to the vehicle according to the data processing result.
  • the UWB-based voice recognition automobile positioning system 400 also includes an audio bus 450, wherein the positioning pulse signal and the voice signal are respectively transmitted to the micro control unit through the audio bus 450; the UWB positioning modules and the voice pickup modules are electrically connected to each other through a connector 460; the number of UWB positioning modules is at least 2, and the number of voice pickup modules is at least 2.
  • the UWB positioning module 410 is mainly used to realize the positioning function.
  • the main principle is as follows: the car key with UWB and BLE (Bluetooth) functions is used as an anchor point.
  • BLE Bluetooth
  • the car key automatically triggers the UWB transmission signal.
  • Multiple UWB-based positioning modules on the vehicle body receive the UWB signal emitted by the car key, which is transmitted to the micro control unit 430 through the audio bus 450 for analysis, and finally confirms the relative position of the car key relative to the car; and issues relevant instructions through the vehicle computer 440.
  • the voice pickup module 420 is mainly used to pick up voice signals outside the vehicle.
  • the main principle is as follows: MEMS/ECM MIC picks up voice signals in the environment, converts electrical signals into digital signals, and transmits them to the micro control unit 430 and the vehicle computer 440 in the form of an audio bus 450 for voice recognition and associated operations.
  • the UWB positioning module 410 and the voice pickup module 420 are powered by the audio bus 450, which can save part of the power supply circuit.
  • the UWB positioning module and the voice pickup module can be integrated into a UWB voice module, which is placed at the left front, left rear, right front, right rear and roof reading light of the car according to the positioning requirements, as shown in Figure 3.
  • Each module (1 to 6) is connected through an audio bus, and the common bus can be A2B or a similar bus with similar functions. The module is powered by the bus and finally transmits the signal to the car computer for processing.
  • UWB communication is performed between the car key and the UWB voice pickup module on the car to determine the relative position of the owner and the car.
  • the voice pickup function is automatically turned on to identify the surrounding voice signals.
  • the owner can unlock the car by issuing voice commands in combination with positioning information. Open specific doors and windows, open the tailgate, turn on the lights, and so on.
  • the UWB-based voice recognition car positioning method and system provided in the present application, through the combination of the UWB positioning module and the voice pickup module, accurately confirms the user's position relative to the car, realizes beamforming noise reduction, and accurately receives user commands, so as to improve the intelligence level of off-vehicle voice recognition, reduce the probability of voice command recognition errors, and efficiently complete the interaction between the car and the user.

Landscapes

  • Engineering & Computer Science (AREA)
  • Human Computer Interaction (AREA)
  • Signal Processing (AREA)
  • Health & Medical Sciences (AREA)
  • Audiology, Speech & Language Pathology (AREA)
  • Computational Linguistics (AREA)
  • Physics & Mathematics (AREA)
  • Acoustics & Sound (AREA)
  • Multimedia (AREA)
  • Quality & Reliability (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Navigation (AREA)
  • Lock And Its Accessories (AREA)

Abstract

本申请提供一种基于UWB的语音识别汽车定位方法及***，其中的方法包括通过车钥匙向汽车所在的位置发送脉冲信号；通过位于汽车内的各个位置的UWB定位模块接收脉冲信号，并标记接收到脉冲信号的时间戳；对脉冲信号以及相应的时间戳进行处理，确定携带车钥匙的用户与汽车的相对位置；根据相对位置，确定相对位置所对应的语音拾取模块；通过DSP对语音拾取模块所拾取的语音信号进行波束成形处理，获取增强后的语音信号；根据增强后的语音信号，控制汽车完成增强后的语音信号相应的指令。利用本申请，能够解决现有的车外语音识别准确率等问题。

Description

基于UWB的语音识别汽车定位方法及***
本申请要求于2022年10月26日提交中国专利局、申请号为202211316954.4，发明名称为“基于UWB的语音识别汽车定位方法及***”的中国专利申请的优先权，其全部内容通过引用结合在本申请中。
技术领域
本申请涉及汽车应用技术领域，更为具体地，涉及一种基于UWB的语音识别汽车定位方法及***。
背景技术
随着UWB(Ultra Wide Band，超宽带)技术的应用普及，汽车市场部分新车型推出了UWB智能车钥匙，结合汽车内部UWB***实现了实时高精度定位，并依据车钥匙及汽车的实时距离控制汽车执行特定的功能。车内UWB定位***通常由3-5个UWB定位模块组成(一个主节点+多个锚点)，多通过CAN(Controller Area Network，控制器域网)、LIN(Local Interconnect Network，局域互联网)总线实现单节点与车机***通信。但是UWB技术主要用于定位，实现的功能比较单一。
当前语音识别功能逐步广泛应用于汽车交互***中，车外语音交互也有一定的应用，但是，发明人意识到，目前的车外语音***仅能识别简单的指令，不适用于复杂的(多名人员同时发声)情况下识别指令，这种多人同时发声下容易导致语音识别错误。
发明内容
鉴于上述问题，本申请的目的是提供一种基于UWB的语音识别汽车定位方法及***，以解决现有的车外语音识别准确率等问题。
本申请提供的基于UWB的语音识别汽车定位方法，包括：
通过车钥匙向汽车所在的位置发送脉冲信号；
通过位于汽车内的各个位置的UWB定位模块接收脉冲信号，并标记接收到脉冲信号的时间戳；
对脉冲信号以及相应的时间戳进行处理，确定携带所述车钥匙的用户与汽车的相对位置；
根据相对位置，确定相对位置所对应的语音拾取模块；
通过DSP对语音拾取模块所拾取的语音信号进行波束成形处理，获取增强后的语音信号；
根据增强后的语音信号，控制汽车完成增强后的语音信号相应的指令。
此外，优选的方案是，在通过车钥匙向汽车所在的位置发送脉冲信号之前，还包括：
将车钥匙与汽车进行蓝牙匹配，汽车定时接收所述车钥匙的蓝牙信号；
根据接收到的蓝牙信号，判断所述车钥匙是否进入到UWB定位模块的工作范围；
当接收到的所述蓝牙信号到达预设值时，确定车钥匙进入到所述UWB定位模块的工作范围；
汽车内的各个UWB模组接收所述车钥匙发出的脉冲信号。
此外，优选的方案是，在对所述脉冲信号以及相应的时间戳进行处理，确定携带所述车钥匙的用户与汽车的相对位置之前，还包括：
通过总线将各个位置的UWB定位模块接收的脉冲信号以及相应的时间戳传输至微控制单元；
通过微控制单元对脉冲信号以及相应的时间戳进行处理，确定携带所述车钥匙的用户与汽车的相对位置。
此外，优选的方案是，采用TOA、TDOA或AOA对脉冲信号以及相应的时间戳进行处理，确定携带所述车钥匙的用户与汽车的相对位置。
此外，优选的方案是，在通过DSP对所述用户与汽车的相对位置所对应的语音拾取模块所拾取的语音信号进行波束成形处理，获取增强后的语音信号之前，包括：
将位于汽车内的各个位置的语音拾取模块获取的语音信号传输到微控制 单元，其中，各个位置的语音拾取模块与各个位置的UWB定位模块对应设置；
通过所述微控制单元对各个位置的语音拾取模块获取的语音分别进行波束成形处理以及信号屏蔽处理。
此外，优选的方案是，通过所述微控制单元对不与用户与汽车的相对应的语音识别模块进行语音信号屏蔽处理。
本申请还提供一种基于UWB的语音识别汽车定位***，包括：UWB定位模块、与UWB定位模块对应设置的语音拾取模块、微控制单元以及车机，其中，
UWB定位模块，用于获取携带车钥匙的用户与汽车的定位脉冲信号，并将定位脉冲信号传输至所述微控制单元；
语音拾取模块，用于获取用户的语音信号，并将语音信号传输至所述微控制单元；
微控制单元，用于对定位脉冲信号以及所述语音信号进行数据处理，并将数据处理结果发送至所述车机；
车机，用于根据数据处理结果向汽车发送控制指令。
此外，优选的方案是，还包括音频总线，其中，
定位脉冲信号、语音信号分别通过音频总线传输至微控制单元。
此外，优选的方案是，UWB定位模块的数量至少为2个，语音拾取模块的数量至少为2个。
此外，优选的方案是，UWB定位模块之间、语音拾取模块之间均通过连接器相互电连接。
从上面的技术方案可知，本申请提供的基于UWB的语音识别汽车定位方法及***，通过UWB定位模块与语音拾取模块相互结合，精准确认用户相对于汽车的位置，实现波束成形降噪，精准的接收用户指令，以提高车外语音识别的智能化程度，降低语音指令识别错误的概率，高效完成车与用户的交互。
为了实现上述以及相关目的，本申请的一个或多个方面包括后面将详细 说明的特征。下面的说明以及附图详细说明了本申请的某些示例性方面。然而，这些方面指示的仅仅是可使用本申请的原理的各种方式中的一些方式。此外，本申请旨在包括所有这些方面以及它们的等同物。
附图说明
通过参考以下结合附图的说明的内容，并且随着对本申请的更全面理解，本申请的其它目的及结果将更加明白及易于理解。在附图中：
图1为根据本申请实施例的基于UWB的语音识别汽车定位方法流程示意图；
图2为根据本申请实施例的基于UWB的语音识别汽车定位流程示意图；
图3为根据本申请实施例的基于UWB的语音识别汽车定位原理示意图；
图4为根据本申请实施例的基于UWB的语音识别汽车定位***逻辑结构示意图。
在所有附图中相同的标号指示相似或相应的特征或功能。
具体实施方式
在下面的描述中，出于说明的目的，为了提供对一个或多个实施例的全面理解，阐述了许多具体细节。然而，很明显，也可以在没有这些具体细节的情况下实现这些实施例。在其它例子中，为了便于描述一个或多个实施例，公知的结构和设备以方框图的形式示出。
在本申请的描述中，需要理解的是，术语“中心”、“纵向”、“横向”、“长度”、“宽度”、“厚度”、“上”、“下”、“前”、“后”、“左”、“右”、“竖直”、“水平”、“顶”、“底”“内”、“外”、“顺时针”、“逆时针”、“轴向”、“径向”、“周向”等指示的方位或位置关系为基于附图所示的方位或位置关系，仅是为了便于描述本申请和简化描述，而不是指示或暗示所指的装置或元件必须具有特定的方位、以特定的方位构造和操作，因此不能理解为对本申请的限制。
针对前述提出的现有的车外语音识别准确率等问题，本申请提供了一种基于UWB的语音识别汽车定位方法及***。
以下将结合附图对本申请的具体实施例进行详细描述。
为了说明本申请提供的基于UWB的语音识别汽车定位方法，图1示出了 根据本申请实施例的基于UWB的语音识别汽车定位方法流程。
如图1所示，本申请提供的基于UWB的语音识别汽车定位方法，包括：
S110：通过车钥匙向汽车所在的位置发送脉冲信号；
S120：通过位于汽车内的各个位置的UWB定位模块接收脉冲信号，并标记接收到脉冲信号的时间戳；
S130：对脉冲信号以及相应的时间戳进行处理，确定携带所述车钥匙的用户与汽车的相对位置；
S140：根据相对位置，确定相对位置所对应的语音拾取模块；
S150：通过DSP对语音拾取模块所拾取的语音信号进行波束成形处理，获取增强后的语音信号；
S160：根据增强后的语音信号，控制汽车完成增强后的语音信号相对应的指令。
在本申请的实施例中，通过UWB定位模块和语音拾取模块相互结合，实现用户对车辆的相对位置的精确定位，实现波束成形降噪，更为精准的接收交互者指令。
在本申请的实施例中，在步骤S110之前，在通过车钥匙向汽车所在的位置发送脉冲信号之前，还包括：
将车钥匙与汽车进行蓝牙匹配，汽车定时接收所述车钥匙的蓝牙信号；
根据接收到的蓝牙信号，判断车钥匙是否进入到UWB定位模块的工作范围；其中，
当接收到的所述蓝牙信号到达预设值时，确定车钥匙进入到所述UWB定位模块的工作范围；
汽车内的各个UWB模组接收所述车钥匙发出的脉冲信号。
在本申请的实施例中，用户使用的车钥匙具有BLE功能，可以定时进行信号广播，汽车接收车钥匙BLE信号，并根据RSSI信号强度判断是否进入UWB定位模块的工作范围。
在对所述脉冲信号以及相应的时间戳进行处理，确定携带所述车钥匙的用户与汽车的相对位置之前，还包括：
通过总线将各个位置的UWB定位模块接收的脉冲信号以及相应的时间戳传输至微控制单元；
通过微控制单元对接收的脉冲信号以及相应的时间戳进行处理，确定用户与汽车的相对位置。
在本申请的实施例中，通过车钥匙与车上UWB定位模块进行UWB通信，确定车主与车子的相对位置。具有UWB+BLE功能的车钥匙作为锚点，当车钥匙端的BLE(Bluetooth Low Energy，蓝牙低能耗)与车身BLE连接，车钥匙自动触发UWB发射信号，车身上的多个基于UWB定位模组接收车钥匙发出的UWB信号，从而确定用户与汽车之间的相对位置。
其中，采用TOA(Time of Arrival，到达时刻)、TDOA(Time Difference of Arrival，到达时间差)或AOA(Angle of Arrival，到达角度测距)对脉冲信号以及相应的时间戳进行处理，确定用户与汽车的相对位置。在具有应用中，根据实际需求选择合适的算法确定用户与汽车的相对位置。
具体地，TDOA是基于各参考基站(以下称为Anchor或Anc)与待定位对象(以下称为Tag)之间的距离之差通过求解非线性双曲方程组来推断待定位对象相对于各参考基站的相对位置的定位方法。由于电磁波飞行速度是已知且恒定的，飞行距离与飞行时间之间是可以相互转换的，而如果发送时间是共同的(或者间隔是已知的)，则飞行时间又进一步与信号到达时间是可以相互转换，因此只需要测量Tag发出的信号到达各Anchor的到达时间之差即可得到对应的距离之差。
在本申请的实施例中，在通过DSP对相对位置所对应的语音拾取模块所拾取的语音信号进行波束成形处理，获取增强后的语音信号之前，包括：
将位于汽车内的各个位置的语音拾取模块获取的语音信号传输到微控制单元，其中，各个位置的语音拾取模块与各个位置的UWB定位模块对应设置；
通过所述微控制单元对各个位置的语音拾取模块获取的语音分别进行波束成形处理以及信号屏蔽处理。
其中，数字信号处理(Digital Signal Processing，简称DSP)，包括语音编码、语音合成、语音识别、语音增强、语音邮件、语音储存等，在本申请的实施例中，采用语音增强技术，通过DSP对语音信号进行增强，获取精准的语音指令。
其中，通过微控制单元对不与用户与汽车的相对应的语音识别模块进行语音信号屏蔽处理。
为了详细说明基于UWB的语音识别汽车定位方法的具体实施例，如图2和图3所示，用户使用的车钥匙具有BLE功能，可以定时进行信号广播，汽车接收车钥匙BLE信号，并根据RSSI信号强度判断是否进入UWB工作范围。
用户进入UWB工作范围，车钥匙定时发送脉冲信号。汽车上通过不同位置的6个模组在接收到车钥匙的脉冲信号后，会添加一个接收脉冲时间点的时间戳。不同位置的模组接收脉冲信号时间戳将会产生差异。1-6#模组(六个UWB定位模块和六个语音拾取模块综合为六个定位语音模组，即一个模组中包括一个UWB定位模块和一个语音拾取模块)的相关数据信息通过总线传递至主节点模组A(微控制单元)。模组A(微控制单元)通过接收到的时间戳，基于TDOA算法，实现用户与人之间的定位，判断用户的实际位置(图3所示，区域1-6)。车钥匙会间歇性发出脉冲信号，供模组A(微控制单元)实时更新定位信息。
根据用户与汽车的相对位置，对与相对位置获取的语音进行语音增强处理，并屏蔽其他位置的语音信号。具体地，模组A(微控制单元)激活语音识别功能，1-6个模组拾取语音，对从1-6个模组中获取的语音信号通过总线(A2B等)获取到的信号进行处理。如图3所示，用户位于区域6，模组A(微控制单元)将对从模组1和2获取到的语音使用DSP进行波束成形算法处理，并对该区域语音信号做增强处理，剩余区域1-5位置获取到的语音信号进行屏蔽处理。此时，如果处于其他位置的行人1-3同时发出误触指令，汽车将不会执行指令，仅执行用户发出的指令。
处理完成后，将语音信号传递至车机，车机根据语音信号的具体内容执行具体的指令。
在基于UWB定位进行语音信号识别，在具有多名人员的情况下，可以准确识别使用者指令，并降低发生不法分子串用车外语音***的功能。同时，在一些特定场合，诸如打开特定位置车窗等场景中，用户不需要使用繁杂的指令与汽车交互，仅需要说出“打开车窗”，通过UWB定位和语音识别，汽车就可以执行更加人性化的操作。
与上述方法相对应，本申请还提供一种基于UWB的语音识别汽车定位***，图4示出了根据本申请实施例的基于UWB的语音识别汽车定位***逻辑结构。
如图4所示，本申请提供的基于UWB的语音识别汽车定位***400，包括：UWB定位模块410、与UWB定位模块410对应设置的语音拾取模块420、微控制单元430以及车机440，其中，
UWB定位模块410，用于获取携带车钥匙的用户与汽车的定位脉冲信号，并将定位脉冲信号传输至所述微控制单元；
语音拾取模块420，用于获取用户的语音信号，并将语音信号传输至所述微控制单元；
微控制单元430，用于对定位脉冲信号以及所述语音信号进行数据处理，并将数据处理结果发送至所述车机；
车机440，用于根据数据处理结果向汽车发送控制指令。
在本申请的实施例中，基于UWB的语音识别汽车定位***400还包括音频总线450，其中，定位脉冲信号、语音信号分别通过音频总线450传输至微控制单元；UWB定位模块之间、语音拾取模块之间均通过连接器460相互电连接；UWB定位模块的数量至少为2个，语音拾取模块的数量至少为2个。
在本申请的实施例中，UWB定位模块410主要用于实现定位功能。主要原理如下：具有UWB和BLE(蓝牙)功能的车钥匙作为锚点，当车钥匙端的BLE与车身BLE连接，车钥匙自动触发UWB发射信号，车身上的多个基于UWB定位模组接收车钥匙发出的UWB信号，通过音频总线450传输至微控制单元430进行解析，最终确认出车钥匙相对于汽车的相对位置；并通过车机440发出相关指令。
语音拾取模块420主要用于拾取车外的语音信号。主要原理如下：MEMS/ECM MIC拾取环境中的语音信号，并将电信号转换为数字信号，以音频总线450的方式传递至微控制单元430和车机440进行语音识别及关联操作。UWB定位模块410和语音拾取模块420通过音频总线450进行供电，可节省部分电源供电电路。
在本申请的一个具体的实施例中，UWB定位模块和语音拾取模块可以综合为UWB语音模组，按照定位需求，如图3所示，分别放置于汽车的左前，左后，右前，右后及车顶阅读灯等位置，各个模组(1至6)之间通过音频总线进行连接，常用总线可以是A2B或者相似具有功能的总线。模组通过总线进行供电，并最终将信号传输至车机进行处理。
在本申请的实施例中，通过车钥匙与车上UWB的语音拾取模组进行UWB通信，确定车主与车子的相对位置。在车主与车子相距一定距离后(可单独设置)，自动打开语音拾取功能，识别周边的语音信号。车主通过发出语音指令，结合定位信息，可以实现解锁。开启特定门窗，开启后尾门，打开车灯等操作。依据UWB定位，通过DSP可以更容易的实现波束成形，更为精准的获取专一指令信息；通过UWB识别在车内还是车外，在车内时自动关闭车外语音指令识别功能，降低非法分子从车外通过语音控制汽车的风险；同时依托语音模组的信号传输线，同步进行UWB模块的控制，减少车内线束及汽车装配复杂度。同时，整体的设计难度及成本相对于UWB/语音分离式设计有所降低。
通过上述实施方式可以看出，本申请提供的基于UWB的语音识别汽车定位方法及***，通过UWB定位模块与语音拾取模块相互结合，精准确认用户相对于汽车的位置，实现波束成形降噪，精准的接收用户指令，以提高车外语音识别的智能化程度，降低语音指令识别错误的概率，高效完成车与用户的交互。
如上参照附图以示例的方式描述了根据本申请提出的基于UWB的语音识别汽车定位方法及***。但是，本领域技术人员应当理解，对于上述本申请所提出的基于UWB的语音识别汽车定位方法及***，还可以在不脱离本申请内容的基础上做出各种改进。因此，本申请的保护范围应当由所附的权利要求书的内容确定。

Claims (10)

  1. 一种基于UWB的语音识别汽车定位方法，其中，包括：
    通过车钥匙向汽车所在的位置发送脉冲信号；
    通过位于汽车内的各个位置的UWB定位模块接收所述脉冲信号，并标记接收到所述脉冲信号的时间戳；
    对所述脉冲信号以及相应的时间戳进行处理，确定携带所述车钥匙的用户与汽车的相对位置；
    根据所述相对位置，确定所述相对位置所对应的语音拾取模块；
    通过DSP对所述语音拾取模块所拾取的语音信号进行波束成形处理，获取增强后的语音信号；
    根据所述增强后的语音信号，控制汽车完成所述增强后的语音信号相对应的指令。
  2. 如权利要求1所述的基于UWB的语音识别汽车定位方法，其中，在通过车钥匙向汽车所在的位置发送脉冲信号之前，还包括：
    将所述车钥匙与汽车进行蓝牙匹配，所述汽车定时接收所述车钥匙的蓝牙信号；
    根据接收到的所述蓝牙信号，判断所述车钥匙是否进入到所述UWB定位模块的工作范围；其中，
    当接收到的所述蓝牙信号到达预设值时，确定所述车钥匙进入到所述UWB定位模块的工作范围，汽车内的各个UWB模组接收所述车钥匙发出的脉冲信号。
  3. 如权利要求1所述的基于UWB的语音识别汽车定位方法，其中，在对所述脉冲信号以及相应的时间戳进行处理，确定携带所述车钥匙的用户与汽车的相对位置之前，还包括：
    通过总线将各个位置的UWB定位模块接收的脉冲信号以及相应的时间戳传输至微控制单元；
    通过所述微控制单元对所述脉冲信号以及相应的时间戳进行处理，确定 携带所述车钥匙的用户与汽车的相对位置。
  4. 如权利要求1所述的基于UWB的语音识别汽车定位方法，其中，采用TOA、TDOA或AOA对所述脉冲信号以及相应的时间戳进行处理，确定携带所述车钥匙的用户与汽车的相对位置。
  5. 如权利要求1所述的基于UWB的语音识别汽车定位方法，其中，在所述通过DSP对所述用户与汽车的相对位置所对应的语音拾取模块所拾取的语音信号进行波束成形处理，获取增强后的语音信号之前，包括：
    将位于汽车内的各个位置的语音拾取模块获取的语音信号传输到微控制单元，其中，所述各个位置的语音拾取模块与所述各个位置的UWB定位模块对应设置；
    通过所述微控制单元对所述各个位置的语音拾取模块获取的语音分别进行波束成形处理以及信号屏蔽处理。
  6. 如权利要求5所述的基于UWB的语音识别汽车定位方法，其中，通过所述微控制单元对不与所述用户与汽车的相对应的语音识别模块进行语音信号屏蔽处理。
  7. 一种基于UWB的语音识别汽车定位***，其中，包括：UWB定位模块、与所述UWB定位模块对应设置的语音拾取模块、微控制单元以及车机，其中，
    所述UWB定位模块，用于获取携带车钥匙的用户与汽车的定位脉冲信号，并将所述定位脉冲信号传输至所述微控制单元；
    所述语音拾取模块，用于获取用户的语音信号，并将所述语音信号传输至所述微控制单元；
    所述微控制单元，用于对所述定位脉冲信号以及所述语音信号进行数据处理，并将数据处理结果发送至所述车机；
    所述车机，用于根据所述数据处理结果向所述汽车发送控制指令。
  8. 如权利要求7所述的基于UWB的语音识别汽车定位***，其中，还包括音频总线，其中，
    所述定位脉冲信号、所述语音信号分别通过所述音频总线传输至所述微控制单元。
  9. 如权利要求7所述的基于UWB的语音识别汽车定位***，其中，所述UWB定位模块的数量至少为2个，所述语音拾取模块的数量至少为2个。
  10. 如权利要求9所述的基于UWB的语音识别汽车定位***，其中，所述UWB定位模块之间、所述语音拾取模块之间均通过连接器相互电连接。
PCT/CN2023/118247 2022-10-26 2023-09-12 基于uwb的语音识别汽车定位方法及*** WO2024087918A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202211316954.4A CN115767719A (zh) 2022-10-26 2022-10-26 基于uwb的语音识别汽车定位方法及***
CN202211316954.4 2022-10-26

Publications (1)

Publication Number Publication Date
WO2024087918A1 true WO2024087918A1 (zh) 2024-05-02

Family

ID=85353723

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/118247 WO2024087918A1 (zh) 2022-10-26 2023-09-12 基于uwb的语音识别汽车定位方法及***

Country Status (2)

Country Link
CN (1) CN115767719A (zh)
WO (1) WO2024087918A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115767719A (zh) * 2022-10-26 2023-03-07 歌尔微电子股份有限公司 基于uwb的语音识别汽车定位方法及***

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20200198580A1 (en) * 2018-12-19 2020-06-25 Ahmad N. Saleh Vehicle passive entry protocol with ultra wide band ranging
CN113844360A (zh) * 2021-09-03 2021-12-28 上海博泰悦臻电子设备制造有限公司 车载人机交互***
US20220335934A1 (en) * 2021-04-19 2022-10-20 GM Global Technology Operations LLC Context-aware signal conditioning for vehicle exterior voice assistant
CN115767719A (zh) * 2022-10-26 2023-03-07 歌尔微电子股份有限公司 基于uwb的语音识别汽车定位方法及***

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20200198580A1 (en) * 2018-12-19 2020-06-25 Ahmad N. Saleh Vehicle passive entry protocol with ultra wide band ranging
US20220335934A1 (en) * 2021-04-19 2022-10-20 GM Global Technology Operations LLC Context-aware signal conditioning for vehicle exterior voice assistant
CN113844360A (zh) * 2021-09-03 2021-12-28 上海博泰悦臻电子设备制造有限公司 车载人机交互***
CN115767719A (zh) * 2022-10-26 2023-03-07 歌尔微电子股份有限公司 基于uwb的语音识别汽车定位方法及***

Also Published As

Publication number Publication date
CN115767719A (zh) 2023-03-07

Similar Documents

Publication Publication Date Title
WO2024087918A1 (zh) 基于uwb的语音识别汽车定位方法及***
US10576932B2 (en) System and methods for vehicle passive keyless entry triggered by smartphone proximity detection
CN109774641B (zh) 车内人员检测方法、雷达和车
CN210072054U (zh) 一种基于rssi值和信号飞行时间的定位***
CN208636719U (zh) 一种基于超声波测距的自动跟随***
US11724667B2 (en) Motor vehicle and method for processing sound from outside the motor vehicle
CN111711921A (zh) 一种寻找车辆的方法、装置、设备及存储介质
CN112258752A (zh) 一种基于智能座舱蓝牙的防手机遗忘车内报警***及方法
CN112584317A (zh) 对移动入口点的触觉引导和导航
CN103680187A (zh) 智能寻车***及智能寻车方法
CN109493641A (zh) 信息处理装置、信息提供***和信息提供方法
CN108205144A (zh) 一种道路作业车辆碰撞预警装置、道路作业车辆及碰撞预警方法
CN101653392A (zh) 具有gps定位、语音识别及蓝牙装置的导盲杖
CN113539259B (zh) 一种基于车辆的语音交流方法和装置
CN108099816A (zh) 一种用于智能出行的车辆语音输入控制方法
CN216719480U (zh) 一种车外语音控制***
CN110040063A (zh) 车辆大灯控制方法、***及车辆
CN216686991U (zh) 基于蓝牙、语音和人脸识别的非接触电梯楼层召唤***
CN208978867U (zh) 一种矿用机车
CN115700401A (zh) 车辆定位方法及相关装置
CN111815962A (zh) 车辆身份的识别方法、装置、主设备和从设备
CN210487978U (zh) 适用于汽车的跟踪识别装置
CN212627974U (zh) 一种车辆远程控制装置
CN114325569B (zh) 一种雷电探测装置
CN117382530A (zh) 车辆开门方法、装置、电子设备、存储介质及程序产品

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: 23881499

Country of ref document: EP

Kind code of ref document: A1