WO2020252743A1 - 一种雷达系统 - Google Patents

一种雷达系统 Download PDF

Info

Publication number
WO2020252743A1
WO2020252743A1 PCT/CN2019/092084 CN2019092084W WO2020252743A1 WO 2020252743 A1 WO2020252743 A1 WO 2020252743A1 CN 2019092084 W CN2019092084 W CN 2019092084W WO 2020252743 A1 WO2020252743 A1 WO 2020252743A1
Authority
WO
WIPO (PCT)
Prior art keywords
data
radar
processing unit
central processing
spectrum
Prior art date
Application number
PCT/CN2019/092084
Other languages
English (en)
French (fr)
Inventor
杨晨
刘劲楠
陈佳民
刘培
Original Assignee
华为技术有限公司
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 华为技术有限公司 filed Critical 华为技术有限公司
Priority to PCT/CN2019/092084 priority Critical patent/WO2020252743A1/zh
Priority to EP19933643.9A priority patent/EP3968054A4/en
Priority to CN201980066312.3A priority patent/CN112805591A/zh
Publication of WO2020252743A1 publication Critical patent/WO2020252743A1/zh
Priority to US17/554,605 priority patent/US20220107390A1/en

Links

Images

Classifications

    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
    • G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
    • G01S7/41—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00 using analysis of echo signal for target characterisation; Target signature; Target cross-section
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
    • G01S13/88—Radar or analogous systems specially adapted for specific applications
    • G01S13/93—Radar or analogous systems specially adapted for specific applications for anti-collision purposes
    • G01S13/931—Radar or analogous systems specially adapted for specific applications for anti-collision purposes of land vehicles
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
    • G01S13/02—Systems using reflection of radio waves, e.g. primary radar systems; Analogous systems
    • G01S13/06—Systems determining position data of a target
    • G01S13/08—Systems for measuring distance only
    • G01S13/32—Systems for measuring distance only using transmission of continuous waves, whether amplitude-, frequency-, or phase-modulated, or unmodulated
    • G01S13/34—Systems for measuring distance only using transmission of continuous waves, whether amplitude-, frequency-, or phase-modulated, or unmodulated using transmission of continuous, frequency-modulated waves while heterodyning the received signal, or a signal derived therefrom, with a locally-generated signal related to the contemporaneously transmitted signal
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
    • G01S13/02—Systems using reflection of radio waves, e.g. primary radar systems; Analogous systems
    • G01S13/06—Systems determining position data of a target
    • G01S13/42—Simultaneous measurement of distance and other co-ordinates
    • G01S13/44—Monopulse radar, i.e. simultaneous lobing
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
    • G01S13/02—Systems using reflection of radio waves, e.g. primary radar systems; Analogous systems
    • G01S13/50—Systems of measurement based on relative movement of target
    • G01S13/58—Velocity or trajectory determination systems; Sense-of-movement determination systems
    • G01S13/583—Velocity or trajectory determination systems; Sense-of-movement determination systems using transmission of continuous unmodulated waves, amplitude-, frequency-, or phase-modulated waves and based upon the Doppler effect resulting from movement of targets
    • G01S13/584—Velocity or trajectory determination systems; Sense-of-movement determination systems using transmission of continuous unmodulated waves, amplitude-, frequency-, or phase-modulated waves and based upon the Doppler effect resulting from movement of targets adapted for simultaneous range and velocity measurements
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
    • G01S13/87—Combinations of radar systems, e.g. primary radar and secondary radar
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
    • G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
    • G01S7/027—Constructional details of housings, e.g. form, type, material or ruggedness
    • G01S7/028—Miniaturisation, e.g. surface mounted device [SMD] packaging or housings
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S7/00—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00
    • G01S7/02—Details of systems according to groups G01S13/00, G01S15/00, G01S17/00 of systems according to group G01S13/00
    • G01S7/40—Means for monitoring or calibrating
    • G01S7/4052—Means for monitoring or calibrating by simulation of echoes
    • G01S7/406—Means for monitoring or calibrating by simulation of echoes using internally generated reference signals, e.g. via delay line, via RF or IF signal injection or via integrated reference reflector or transponder
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
    • G01S13/88—Radar or analogous systems specially adapted for specific applications
    • G01S13/93—Radar or analogous systems specially adapted for specific applications for anti-collision purposes
    • G01S13/931—Radar or analogous systems specially adapted for specific applications for anti-collision purposes of land vehicles
    • G01S2013/9327—Sensor installation details
    • G01S2013/93271—Sensor installation details in the front of the vehicles
    • G—PHYSICS
    • G01—MEASURING; TESTING
    • G01S—RADIO DIRECTION-FINDING; RADIO NAVIGATION; DETERMINING DISTANCE OR VELOCITY BY USE OF RADIO WAVES; LOCATING OR PRESENCE-DETECTING BY USE OF THE REFLECTION OR RERADIATION OF RADIO WAVES; ANALOGOUS ARRANGEMENTS USING OTHER WAVES
    • G01S13/00—Systems using the reflection or reradiation of radio waves, e.g. radar systems; Analogous systems using reflection or reradiation of waves whose nature or wavelength is irrelevant or unspecified
    • G01S13/88—Radar or analogous systems specially adapted for specific applications
    • G01S13/93—Radar or analogous systems specially adapted for specific applications for anti-collision purposes
    • G01S13/931—Radar or analogous systems specially adapted for specific applications for anti-collision purposes of land vehicles
    • G01S2013/9327—Sensor installation details
    • G01S2013/93272—Sensor installation details in the back of the vehicles

Definitions

  • This application relates to the field of radar technology, in particular to a radar system.
  • High-resolution vehicle-mounted radar is an indispensable and important part of automatic driving, used to measure target distance and speed.
  • radar sensors need to provide stronger environmental perception capabilities.
  • Increasing the scale of the radar sensor array antenna is a typical method to improve the sensor's ability to measure the environment.
  • more transceiver antennas can enhance the radar sensor's ability to distinguish the target angle.
  • large-scale antenna arrays bring stronger data throughput requirements, more data storage requirements, and greater computing power requirements. Therefore, subsequent signal processing and data processing capabilities also need to match the gradually increasing sensor specifications.
  • the position where the radar can be installed on the vehicle body and the corresponding installation size have not continued to increase.
  • the antenna array scale of the radar sensor increases, the number of receiving antennas increases significantly.
  • the data processing of multiple radar sensors is If the central processing unit is used for unified processing, it will bring huge challenges to the computing power of the central processing unit; if the processor at the radar sensor is processed and sent to the central processing unit for summary, the processor at the radar sensor It will soon enter the bottleneck of processing power, buffering capacity and data throughput; these shortcomings will restrict the adaptability of radar systems to large-scale antenna array radar sensors.
  • the embodiment of the present invention provides a radar system, which can preprocess the data acquired by the radar sensor through a microprocessor to reduce the processing pressure of the subsequent processor, thereby improving the data processing capability of the radar system.
  • the first aspect of the present application provides a radar system, including at least one radar sensor.
  • the first radar sensor in the at least one radar sensor includes a data merging module and M radar chips, wherein each radar chip includes a first radio frequency.
  • Unit and the first microprocessor where M is an integer greater than 1, where the first microprocessor may be a radar-signal processing unit (RPU), and the radar single chip may be composed of separate modules
  • the circuit structure may also be a system-on-chip (SoC);
  • the first radio frequency unit is used to receive echo data, where the echo data can be the chirp transmitted by the first radio frequency unit Echo data;
  • the first microprocessor is used to preprocess the echo data to obtain first data;
  • the data merging module is used to merge the first data output by the M radar chips to obtain The merged data of the first radar sensor;
  • the data merging module is also used to transmit the merged data, the merged data is used to obtain the second data of the first radar sensor after post-processing, the second data of each
  • each radar chip in the first radar sensor includes a first radio frequency unit and a first microprocessor.
  • the first microprocessor responds to the response obtained by the first radio frequency unit.
  • the first data is obtained, and the first data is sent to the data merging module.
  • the data merging module is connected to all the radar chips.
  • the data merging module receives the first data sent by each radar chip. After that, all the first data is merged into the merged data, and the merged data is the preprocessed data of the first radar sensor. Furthermore, only the merged data needs to be further processed later to obtain the first radar.
  • the second data of the sensor, the radar system contains multiple radar sensors, and the second data of all the radar sensors is finally used to generate the point cloud data of the radar system.
  • the acquired echo data obtained by the first radio frequency unit is not completely processed by the microprocessor of the radar sensor, but is preprocessed by the microprocessor in the radar sensor and sent to the next Post-processing is performed in one link.
  • the radar sensor and the next level of equipment share the computing power pressure of the echo data processing, thus solving the problem of processing at the radar sensor end when the antenna array of the radar sensor increases.
  • the device will soon enter the bottleneck problem of processing capacity, buffering capacity and data throughput capacity, which improves the adaptability of radar systems to large-scale antenna array radar sensors.
  • the radar system further includes a central processing unit, and the central processing unit may be a central radar-signal processing unit (central RPU).
  • the merging module is also used to transmit the merged data, including: the data merging module is used to transmit the merged data to the central processing unit; the central processing unit is used to post-process the merged data to obtain the first radar sensor Two data.
  • the data merging module transmits the merged data to the central processing unit, and the central processing unit preprocesses the merged data to obtain the second data of the first radar sensor, because the merged data has already passed through the first radar sensor.
  • the pre-processing of the microprocessor therefore, the central processing unit only needs less computing power to calculate the second data, thereby reducing the processing pressure of the central processing unit.
  • the combined data transmitted is pre-processed, it is also Improve the data throughput capacity between the data merging module and the central processing unit.
  • the signal transmission bit width between the data merging module and the central processing unit is greater than or equal to the combined data of the first radar sensor ,
  • the data merging module is used for merging the first data output by each of the M radar chips, including: the data merging module is used for performing the first data output by the M radar chips Bale.
  • the signal transmission bit width between the data merging module and the central processing unit is greater than or equal to the bit width of the combined data of the first radar sensor, that is, the transmission channel between the data merging module and the central processing unit can be complete
  • the combined data of the first radar sensor is transmitted to the ground. Therefore, the data combining module only needs to package the first data output by each of the M radar chips, and then send the combined data to the central processing unit.
  • the signal transmission bit width between the data merging module and the central processing unit is smaller than the bit width of the merged data of the first radar sensor.
  • the data merging module is used to merge the first data output by the M radar chips, including: the data merging module is used to merge the first data output by the M radar chips from the parallel data Converted into serial data; the data merging module is used for buffering the serial data; the data merging module is used for transmitting the merged data to the central processing unit, including: the data merging module is used for N times to the central processing unit When transmitting the serial data, the N is an integer greater than 1.
  • the data merging module when the signal transmission bit width between the data merging module and the central processing unit is smaller than the bit width of the combined data of the first radar sensor, the data merging module cannot transmit the combined data to the central processing unit at one time through the transmission channel. Therefore, the data merging module needs to convert the merged data from parallel data to serial data and buffer it, and then transmit it to the central processing unit in stages, so that when the transmission bit width is smaller than the merged data bit width, the data merging module can transmit to the central processing unit Complete consolidated data.
  • the first microprocessor is configured to preprocess the echo data to obtain first data, including: The first microprocessor is used for performing ranging processing on the echo data to obtain ranging data; the data merging module is used for merging the first data output by each of the M radar chips to obtain the first radar
  • the merged data of the sensor includes: the data merging module is used for merging the ranging data output by each of the M radar single chips to obtain the merged data of the first radar sensor; the central processing unit is used for performing the merged data
  • the post-processing to obtain the second data of the first radar sensor includes: the central processing unit is used to perform velocity measurement processing and angle measurement processing on the combined data to obtain velocity measurement data and angle measurement data of the first radar sensor.
  • the first microprocessor at one end of the radar sensor is used to perform ranging processing on the echo data to complete the preprocessing operation, and then the central processing unit performs velocity measurement processing and angle measurement on the ranging processed data Processing, so that the first microprocessor and the central processing unit share the pressure on the computing power of the echo data, and improve the data processing ability.
  • the amount of data is reduced, thereby increasing Data throughput capacity.
  • the first microprocessor is configured to perform ranging processing on the echo data to obtain ranging data, including: A microprocessor is used to perform data zeroing on the echo data; the first microprocessor is also used to perform windowing processing on the echo data after data zeroing; the first microprocessor is also used to add The echo data after the window is subjected to fast Fourier transform to obtain first spectrum data.
  • the first spectrum data includes at least one first spectrum, wherein each single frequency in the first spectrum corresponds to a detection point Distance data, the detection point is a point in the point cloud information.
  • the first microprocessor performs zero padding on the echo data so that the echo data meets the processing requirements, performs windowing processing on the echo data after zero padding, and then performs fast Fourier transform to obtain the first A spectrum data, by interpreting the first spectrum data, where each single frequency in the first spectrum corresponds to the distance data of a detection point, thus obtaining the distance information of all detection points of the radar chip, and the subsequent central processing During the processing of the radar, the detection points of all radar sensors constitute the point cloud of the radar system.
  • the central processing unit is used to perform velocity measurement processing and angle measurement processing on the combined data to obtain velocity measurement data of the first radar sensor
  • the angle measurement data includes: the central processing unit is used to obtain the speed measurement data of each detection point according to the distance data corresponding to each single frequency in the first spectrum data.
  • the central processing unit is used to perform velocity measurement processing and angle measurement processing on the combined data to obtain velocity measurement data of the first radar sensor And angle measurement data, including:
  • the central processing unit is configured to obtain second spectrum data, the second spectrum data being spectrum data obtained after the first spectrum data is subjected to ranging processing and speed measurement processing, wherein the second spectrum data includes at least one second spectrum,
  • the second frequency spectrum is a frequency spectrum obtained after the first frequency spectrum is subjected to ranging processing and speed measurement processing;
  • the central processing unit is configured to accumulate all the second spectra in the second spectrum data, where the accumulation may be coherent accumulation or non-coherent accumulation;
  • the central processing unit is used to perform fast Fourier transform on the detection points at the same position of each second frequency spectrum, so as to obtain the angle measurement data of each detection point.
  • the first microprocessor performs ranging processing on the echo data to obtain the first spectrum data
  • the central processing unit performs speed measurement processing on the first spectrum data to obtain the second spectrum data
  • the central processing unit performs the range measurement on the second spectrum data. All the second frequency spectra in the data are accumulated, and the fast Fourier transform is performed on the detection points at the same position of the second frequency spectrum, so as to obtain the angle measurement data of each detection point; in this way, the distance measurement, speed measurement and speed measurement of all detection points are performed. Angle measurement to achieve tracking of all detection points, and then select the target to be detected from the tracked detection points, which requires a large amount of calculation.
  • the first microprocessor and the central processing unit can work together. The computing power required to support this working method.
  • the central processing unit is used to perform velocity measurement processing and angle measurement processing on the combined data to obtain velocity measurement data of the first radar sensor And angle measurement data, including: the central processing unit is used to obtain second spectrum data, and the second spectrum data is spectrum data obtained after ranging processing and speed measurement processing of the first spectrum data, wherein, in the second spectrum data At least one second frequency spectrum is included, and the second frequency spectrum is a frequency spectrum obtained after the first frequency spectrum is subjected to ranging processing and speed measurement processing; the central processing unit is used for accumulating all the second frequency spectrums in the second spectrum data, Wherein, the accumulation may be coherent accumulation or non-coherent accumulation; the central processing unit is used to perform a two-dimensional constant false alarm rate on the accumulated second spectrum data to obtain the first target detection point from the detection point, The first target detection point is a part of the detection points; in all the second frequency spectra of the second spectrum data, the central processing unit is used to perform fast Fourier transform on
  • the central processing unit performs speed measurement processing on the first spectrum data to obtain second spectrum data, accumulates all the second spectrums in the second spectrum data, and then performs a two-dimensional constant on the accumulated second spectrum data.
  • False alarm rate to extract the part of the detection points that need to be tracked as the first target detection point, and then perform fast Fourier transform on each first target detection point to obtain the angle measurement of each first target detection point Data, this method is first detected, and then the points that need to be tracked are screened out through the two-dimensional constant false alarm rate for tracking, which saves computing power.
  • the first radio frequency unit is also used to send a first chirp signal, and the first chirp signal is used to perform Detection; the first radio frequency unit for receiving echo data, including: the first radio frequency unit for receiving a second chirp signal, the second chirp signal for the first radio frequency unit after sending the first chirp signal The returned chirp signal.
  • the first chirp signal transmitted by the first radio frequency unit bounces after touching an obstacle, so that the first radio frequency unit can receive the second chirp signal, thereby enabling detection.
  • the first microprocessor is configured to preprocess the echo data to obtain first data, including: the first The microprocessor is used for preprocessing the second chirp signal to obtain ranging data; the data merging module is used for merging the ranging data output by each of the M radar chips to obtain the first radar sensor
  • the combined data includes: the data merging module is used to output a second digital chirp signal from each radar chip of the M radar chips to measure the distance measurement data of the respective output of the M radar chips.
  • the range data is merged to obtain the merged data of the first radar sensor; the central processor is used for post-processing the merged data to obtain the second data of the first radar sensor, including:
  • the central processing unit is used to perform speed measurement processing and angle measurement processing on the second digital chirp signal when the data amount of the second digital chirp signal corresponding to the ranging data reaches one frame.
  • the distance data is sent to the central processing unit; since multiple sets of distance measurement data are required to perform speed measurement, when the data amount of the second digital chirp signal corresponding to the distance measurement data reaches one frame, the central processing unit will Digital chirp signal is processed for speed measurement and angle measurement.
  • the first microprocessor is configured to preprocess the echo data to obtain the first data, including:
  • the first microprocessor is used for down-sampling the X second digital chirp signals received by the first radio frequency unit to obtain Y third digital chirp signals, where X and Y are both positive and greater than 1. Integer, where Y is less than X;
  • the data merging module is used to merge the first data output by each of the M radar chips to obtain the merged data of the first radar sensor, including:
  • the data merging module is used for merging the third digital chirp signals output by each of the M radar chips to obtain merged data of the first radar sensor;
  • the central processor is used for post-processing the combined data to obtain the second data of the first radar sensor, including:
  • the central processor is used for post-processing the third digital chirp signal in the combined data to obtain the second data of the first radar sensor.
  • the first microprocessor when the sampling amount of the first radio frequency unit is greater than the amount required by the radar system, the first microprocessor needs to down-sample the second digital chirp signal received by the first radio frequency unit to compress the first radio frequency unit.
  • the amount of data received by the radio frequency unit meets the working requirements of the radar system.
  • the first microprocessor is used to preprocess the echo data to obtain The first data, including;
  • the first microprocessor is used to compress the echo data to obtain first compressed data
  • the data merging module is used to merge the first data output by each of the M radar chips to obtain the merged data of the first radar sensor, including:
  • the data merging module is configured to merge the first compressed data output by each of the M radar chips to obtain the compressed and merged data of the first radar sensor;
  • the central processing unit is used for post-processing the combined data, including:
  • the central processing unit is used to decompress the compressed and combined data.
  • the ranging data is further compressed, so that the data merging module merges and transmits the compressed data, thereby reducing the data merging module and the central
  • the pressure of data transmission between processors improves data throughput.
  • a radar system which can be applied to a vehicle-mounted radar for autonomous driving, such as a vehicle-mounted millimeter wave radar, including at least one radar sensor, and the first radar sensor in the at least one radar sensor includes a data merging module and M radar chips, where each radar chip includes a first radio frequency unit and a first microprocessor, where M is an integer greater than 1; the first radio frequency unit is used to receive echo data; the first microprocessor The device is used to preprocess the echo data to obtain the first data; the data merging module is used to merge the first data output by each of the M radar chips to obtain the combined data of the first radar sensor; The data merging module is also used to transmit the merged data, the merged data is used to obtain the second data of the first radar sensor after post-processing, and the respective second data of the at least one radar sensor is used to generate the points of the radar system Cloud information.
  • a vehicle-mounted radar for autonomous driving such as a vehicle-mounted millimeter wave radar, including at least one radar sensor
  • the first microprocessor and the central processing unit share tasks, so that the system can be compatible with radar sensors of a larger antenna array, and can adapt to different system parameters and algorithm processing procedures; the microprocessor and the central processing unit The burden of computing power and hardware resources are reduced accordingly, and resource allocation is more reasonable. At the same time, because the microprocessor and the central processing unit work together, the processing efficiency is improved.
  • a second aspect of the present application provides a radar system, including at least one radar sensor, and the first radar sensor in the at least one radar sensor includes a first microprocessor and at least one first radio frequency unit; wherein, the first microprocessor It can be a radar-signal processing unit (RPU); the at least one first radio frequency unit is used to receive at least one echo data, where the echo data can be a chirp signal transmitted by the first radio frequency unit ( chirp) echo data; the first microprocessor is used to preprocess the at least one echo data to obtain first data; the first microprocessor is also used to transmit the first data, the first data Used to obtain the second data of the first radar sensor after post-processing, and the respective second data of the at least one radar sensor is used to generate point cloud information of the radar system.
  • the first microprocessor It can be a radar-signal processing unit (RPU); the at least one first radio frequency unit is used to receive at least one echo data, where the echo data can be a chirp signal transmitted by the first radio frequency unit
  • the first radar sensor includes a first microprocessor and at least one first radio frequency unit.
  • the first microprocessor preprocesses the echo data obtained by the first radio frequency unit to obtain the first data, and The first data is transmitted to the next-level processor, and then only the first data needs to be further processed to obtain the second data of the first radar sensor.
  • the radar system contains multiple radar sensors.
  • the second data is finally used to generate point cloud data of the radar system.
  • the acquired echo data obtained by the first radio frequency unit is not completely processed by the first microprocessor of the radar sensor, but is preprocessed by the first microprocessor of the radar sensor. , Send it to the next link for post-processing.
  • the radar sensor and the next level of equipment share the computing power pressure of the echo data processing, which solves the problem that the radar sensor antenna array size increases.
  • the processor on the sensor side will soon enter the bottleneck problem of processing capacity, cache capacity and data throughput capacity, which improves the adaptability of the radar system to large-scale antenna array radar sensors.
  • the radar system further includes a central processing unit, and the central processing unit may be a central radar-signal processing unit (central RPU), which The first microprocessor is also used to transmit the first data, including: the first microprocessor is used to transmit the first data to the central processing unit; the central processing unit is used to post-process the first data to obtain The second data of the first radar sensor.
  • central RPU central radar-signal processing unit
  • the first microprocessor transmits the first data to the central processing unit, and the central processing unit preprocesses the first data to obtain the second data of the first radar sensor.
  • the data has been pre-processed by the first microprocessor. Therefore, the central processing unit only needs less computing power to calculate the second data, thereby reducing the processing pressure of the central processing unit.
  • the data is preprocessed, which also improves the data throughput between the first microprocessor and the central processing unit.
  • the first microprocessor is configured to preprocess the echo data to obtain first data, including: the first The microprocessor is used to perform ranging processing on the echo data to obtain ranging data; the central processor is used to perform post-processing on the first data to obtain the second data of the first radar sensor, including: the central The processor is used to perform velocity measurement processing and angle measurement processing on the first data to obtain velocity measurement data and angle measurement data of the first radar sensor.
  • the first microprocessor is used to perform distance measurement processing on the echo data to complete the preprocessing operation, and then the central processing unit performs speed measurement processing and angle measurement processing on the data after the distance measurement processing, so that The first microprocessor and the central processing unit share the computing power pressure on the echo data and improve the data processing capability.
  • the data volume becomes smaller, thereby improving the data throughput capacity.
  • the first microprocessor is configured to perform ranging processing on the echo data to obtain ranging data, including: A microprocessor is used to perform data zeroing on the echo data; the first microprocessor is also used to perform windowing processing on the echo data after data zeroing; the first microprocessor is also used to add The echo data after the window is subjected to fast Fourier transform to obtain first spectrum data.
  • the first spectrum data includes at least one first spectrum, wherein each single frequency in the first spectrum corresponds to a detection point Distance data, the detection point is a point in the point cloud information.
  • the first microprocessor performs zero padding on the echo data so that the echo data meets the processing requirements, performs windowing processing on the echo data after zero padding, and then performs fast Fourier transform to obtain the first A spectrum data, by interpreting the first spectrum data, where each single frequency in the first spectrum corresponds to the distance data of a detection point, thus obtaining the distance information of all detection points of the radar chip, and the subsequent central processing During the processing of the radar, the detection points of all radar sensors constitute the point cloud of the radar system.
  • the central processing unit is used to perform velocity measurement processing and angle measurement processing on the first data to obtain the velocity measurement of the first radar sensor
  • the data and the angle measurement data include: the central processing unit is used to obtain the speed measurement data of each detection point according to the distance data corresponding to each single frequency in the first spectrum data.
  • the central processing unit is used to perform velocity measurement processing and angle measurement processing on the first data to obtain the velocity measurement of the first radar sensor
  • the data and the angle measurement data include: the central processing unit is used to obtain second spectrum data, the second spectrum data being the spectrum data obtained by the first spectrum data after ranging processing and speed measurement processing, wherein the second spectrum data Contains at least one second frequency spectrum, and the second frequency spectrum is a frequency spectrum obtained after the first frequency spectrum is subjected to ranging processing and speed measurement processing; the central processing unit is used for accumulating all second frequency spectrums in the second spectrum data, Wherein, the accumulation can be coherent accumulation or non-coherent accumulation; in all the second spectra of the second spectrum data, the central processing unit is used to perform fast Fourier analysis on the detection points at the same position of each second spectrum Transform to obtain the angle measurement data of each detection point.
  • the first microprocessor performs ranging processing on the echo data to obtain the first spectrum data
  • the central processing unit performs speed measurement processing on the first spectrum data to obtain the second spectrum data
  • the central processing unit performs the range measurement processing on the second spectrum data. All the second frequency spectra in the data are accumulated, and the fast Fourier transform is performed on the detection points at the same position of the second frequency spectrum, so as to obtain the angle measurement data of each detection point; in this way, the distance measurement, speed measurement and speed measurement of all detection points are performed. Angle measurement to achieve tracking of all detection points, and then select the target to be detected from the tracked detection points, which requires a large amount of calculation.
  • the first microprocessor and the central processing unit can work together. The computing power required to support this working method.
  • the central processing unit is used to perform velocity measurement processing and angle measurement processing on the first data to obtain the velocity measurement of the first radar sensor
  • the data and the angle measurement data include: the central processing unit is used to obtain second spectrum data, the second spectrum data being the spectrum data obtained by the first spectrum data after ranging processing and speed measurement processing, wherein the second spectrum data Contains at least one second frequency spectrum, and the second frequency spectrum is a frequency spectrum obtained after the first frequency spectrum is subjected to ranging processing and speed measurement processing; the central processing unit is used to accumulate all the second frequency spectrums in the second spectrum data , Where the accumulation may be coherent accumulation or non-coherent accumulation; the central processing unit is used to perform a two-dimensional constant false alarm rate on the accumulated second spectrum data to obtain the first target detection point from the detection point , The first target detection point is a part of the detection points; in all the second spectra of the second spectrum data, the central processing unit is used to perform fast Fourier analysis on each first target
  • the central processing unit performs speed measurement processing on the first spectrum data to obtain second spectrum data, accumulates all the second spectrums in the second spectrum data, and then performs a two-dimensional constant on the accumulated second spectrum data.
  • False alarm rate to extract the part of the detection points that need to be tracked as the first target detection point, and then perform fast Fourier transform on each first target detection point to obtain the angle measurement of each first target detection point Data, this method is first detected, and then the points that need to be tracked are screened out through the two-dimensional constant false alarm rate for tracking, which saves computing power.
  • the first radio frequency unit is also used to send a first chirp signal, and the first chirp signal is used for detection;
  • the first radio frequency unit is used to receive echo data, including: the first radio frequency unit is used to receive a second chirp signal, and the second chirp signal is the linearity returned after the first radio frequency unit sends the first chirp signal. FM signal.
  • the first chirp signal transmitted by the first radio frequency unit bounces after touching an obstacle, so that the first radio frequency unit can receive the second chirp signal, thereby enabling detection.
  • the first microprocessor is configured to preprocess the echo data to obtain first data, including: the first The microprocessor is used for preprocessing the second chirp signal to obtain ranging data; the central processing unit is used for postprocessing the first data to obtain the second data of the first radar sensor, including: The central processing unit is used to perform speed measurement processing and angle measurement processing on the second digital chirp signal when the data amount of the second digital chirp signal corresponding to the ranging data has accumulated to one frame.
  • the distance data is sent to the central processing unit; since multiple sets of distance measurement data are required to perform speed measurement, when the data amount of the second digital chirp signal corresponding to the distance measurement data reaches one frame, the central processing unit will The linear frequency modulation signal is processed for speed measurement and angle measurement.
  • the first microprocessor is configured to preprocess the echo data to obtain first data, including: the first The microprocessor is used for down-sampling the X second digital chirp signals received by the first radio frequency unit to obtain Y third digital chirp signals, where X and Y are both positive integers greater than 1, where , Y is less than X; the central processing unit is used for post-processing the first data to obtain the second data of the first radar sensor, including: the central processing unit is used for the third digital chirp in the first data The signal is processed for distance measurement, speed measurement and angle measurement to obtain distance measurement data, speed measurement data and angle measurement data.
  • the first microprocessor when the sampling amount of the first radio frequency unit is greater than the amount required by the radar system, the first microprocessor needs to down-sample the second digital chirp signal received by the first radio frequency unit to compress the first radio frequency unit.
  • the amount of data received by the radio frequency unit meets the working requirements of the radar system.
  • the first microprocessor is configured to preprocess the echo data to obtain the first data, including;
  • the first microprocessor is used for compressing the echo data to obtain first compressed data;
  • the central processing unit is used for post-processing the combined data, including: the central processing unit is used for decompressing the compressed combined data .
  • the ranging data is further compressed, so that the first microprocessor transmits the compressed data to the central processing unit, thereby reducing the number of The pressure of data transmission between a microprocessor and central processing unit increases the data throughput capacity.
  • a radar system which can be applied to a vehicle-mounted radar for autonomous driving, such as a vehicle-mounted millimeter wave radar, including at least one radar sensor, and the first radar sensor in the at least one radar sensor includes a data merging module and M radar chips, where each radar chip includes a first radio frequency unit and a first microprocessor, where M is an integer greater than 1; the first radio frequency unit is used to receive echo data; the first microprocessor The device is used to preprocess the echo data to obtain the first data; the data merging module is used to merge the first data output by each of the M radar chips to obtain the combined data of the first radar sensor; The data merging module is also used to transmit the merged data, the merged data is used to obtain the second data of the first radar sensor after post-processing, and the respective second data of the at least one radar sensor is used to generate the points of the radar system Cloud information.
  • a vehicle-mounted radar for autonomous driving such as a vehicle-mounted millimeter wave radar, including at least one radar sensor
  • the first microprocessor and the central processing unit share tasks, so that the system can be compatible with radar sensors of a larger antenna array, and can adapt to different system parameters and algorithm processing procedures; the microprocessor and the central processing unit The burden of computing power and hardware resources are reduced accordingly, and resource allocation is more reasonable. At the same time, because the microprocessor and the central processing unit work together, the processing efficiency is improved.
  • Figure 1 is a system architecture diagram of a distributed radar signal processing unit currently used
  • Figure 2 shows the current system architecture of a single central radar signal processing unit
  • Figure 3 is a system architecture diagram of a radar system in an embodiment of the present invention.
  • FIG. 4 is a schematic diagram of an implementation manner of a radar sensor of a radar system in an embodiment of the present invention
  • FIG. 5 is a schematic diagram of another implementation manner of the radar sensor of the radar system in the embodiment of the present invention.
  • FIG. 6 is a schematic diagram of another implementation manner of the radar sensor of the radar system in the embodiment of the present invention.
  • Fig. 7 is a schematic diagram of an implementation manner of a radar system in an embodiment of the present invention.
  • FIG. 8 is a schematic diagram of another implementation manner of a radar system in an embodiment of the present invention.
  • Figure 9 is a schematic diagram of a radar sensor
  • Fig. 10 is a flow chart of radar system data processing in an embodiment of the present invention.
  • FIG. 11 is a schematic diagram of performing angle measurement processing on the second frequency spectrum when the radar system in the embodiment of the present invention processes data.
  • the embodiment of the present invention provides a radar system, which can adjust the division of labor of the radar system by processing radar data in a multi-level domain, and improve the processing capability, data caching capability, and data throughput capability of the radar system.
  • At least one (item) refers to one or more, and “multiple” refers to two or more.
  • “And/or” is used to describe the association relationship of associated objects, indicating that there can be three types of relationships, for example, “A and/or B” can mean: only A, only B, and both A and B , Where A and B can be singular or plural.
  • the character “/” generally indicates that the associated objects are in an “or” relationship.
  • the following at least one item (a)” or similar expressions refers to any combination of these items, including any combination of a single item (a) or plural items (a).
  • At least one (a) of a, b or c can mean: a, b, c, "a and b", “a and c", “b and c", or "a and b and c" ", where a, b, and c can be single or multiple.
  • High-resolution vehicle-mounted radar is an indispensable and important part of automatic driving, used to measure target distance and speed.
  • radar sensors need to provide stronger environmental perception capabilities.
  • Increasing the scale of the radar sensor array antenna is a typical method to improve the sensor's ability to measure the environment.
  • more transceiver antennas can enhance the radar sensor's ability to distinguish the target angle.
  • large-scale antenna arrays bring stronger data throughput requirements, more data storage requirements, and greater computing power requirements. Therefore, subsequent signal processing and data processing capabilities also need to match the gradually increasing sensor specifications.
  • the position where the radar can be installed on the vehicle body and the corresponding installation size have not continued to increase. As the antenna array of the radar sensor increases, the number of receiving antennas has increased significantly.
  • Fig. 1 is a system architecture diagram of the currently used distributed radar signal processing unit. As shown in Fig. 1, 6 radars are arranged around the vehicle body 100 (there may be more than one, and the comparison is not limited). It is a forward radar sensor 101, a backward radar sensor 102, and a four-angle radar sensor 103. Each radar sensor includes a digital front end (DFE) and a sensor radar-signal processing unit (Sensor radar-signal processing unit) Sensor RPU Two parts. Among them, the DFE includes a transceiver antenna and a corresponding radio frequency front-end circuit, and transmits the converted digital data to the Sensor RPU for processing through an analog-digital converter (ADC).
  • ADC analog-digital converter
  • the Sensor RPU performs millimeter wave radar signal processing, which usually includes signal processing such as ranging, speed measurement, and angle measurement, and may also include high-level data processing such as point cloud data aggregation and moving target tracking. After the respective data processing of each radar, the result information is transmitted in the form of a digital signal to the central radar signal processing unit 104 (central RPU) for subsequent processing tasks such as multi-radar data fusion.
  • signal processing such as ranging, speed measurement, and angle measurement
  • high-level data processing such as point cloud data aggregation and moving target tracking.
  • the architecture of Figure 1 is used to build a vehicle-mounted millimeter-wave radar system.
  • the scale of the radar sensor antenna array gradually increases, the amount of data acquisition and data transmission bandwidth gradually increase, and are limited by volume, weight and power consumption, each radar sensor will soon enter To the bottleneck of processing power, cache capacity and data throughput capacity. Therefore, it is difficult to adapt to the radar sensor processing requirements of a larger antenna array scale, and it is difficult to be compatible with more flexible system parameter design.
  • Figure 2 is the current system architecture of a single central RPU.
  • 6 radars (or more than one, but the comparison is not limited) are arranged around the car body 200, which are forward radars.
  • the sensor 201, the backward radar sensor 202, and the four-corner radar sensor 203 are each radar sensor.
  • Each radar only contains an analog front end (AFE), and the AFE contains a transceiver antenna and a radio frequency front end circuit.
  • AFE analog front end
  • the analog signal is transmitted to the central radar signal processing unit 204 (central RPU).
  • the central RPU204 is equipped with multiple ADCs in the processor corresponding to multiple analog signal paths of each radar. After the analog signal is converted by ADC, the radar digital signal processing and high-level data fusion processing are performed inside the central RPU204.
  • the architecture of Figure 2 is used to build a vehicle-mounted millimeter-wave radar system.
  • the number of receiving antennas increases significantly, which will result in a substantial increase in the number of analog signal channels output by each radar sensor.
  • dozens or even hundreds of analog signal receiving ports and corresponding ADCs must be configured at one end of the central RPU204, which will result in a substantial increase in the chip area of the central RPU204 and the size of the corresponding processing board.
  • the central RPU204 since the central RPU204 is responsible for signal processing and data fusion tasks, it poses a huge challenge to the processor's computing power.
  • the central processing unit if the central processing unit is used for unified processing, it will bring a huge challenge to the computing power of the central processing unit; if the processing power of the radar sensor is used, it will be sent to the central processing unit.
  • the processor at one end of the radar sensor will quickly enter the bottleneck of processing capacity, cache capacity and data throughput capacity; these shortcomings will restrict the adaptability of the radar system to large-scale antenna array radar sensors.
  • the embodiments of the present application provide a radar system, which adjusts the division of labor of the radar system by processing radar data in a multi-level domain, and improves the processing capability, data buffer capability, and data throughput capability of the radar system.
  • the use scenarios of the radar system provided in this application are not limited to vehicle-mounted radars, but can also be used in other radar systems and other electronic processing systems with similar topologies.
  • the flexible configuration of processing tasks and processing modes makes the system more compatible.
  • kind of peripheral configuration With standardized, miniaturized distributed processing unit and powerful central processing unit to achieve the soft upgrade of the system.
  • the radar system provided by the present application includes at least one radar sensor.
  • the first radar sensor in the at least one radar sensor includes a data merging module and M radar chips.
  • Each radar chip includes a first radio frequency unit and a first radar chip.
  • the first radio frequency unit is used to receive echo data
  • the first microprocessor is used to preprocess the echo data to obtain first data
  • the data merging module is used to merge the first data output by each of the M radar chips to obtain the merged data of the first radar sensor;
  • the data merging module is also used to transmit the merged data, the merged data is used to obtain the second data of the first radar sensor after post-processing, and the respective second data of the at least one radar sensor is used to generate the points of the radar system Cloud information.
  • the data transmitted by the data merging module may be sent to the central processing unit.
  • the lower level processor is the central processing unit as an example for description, but this embodiment of the application does not limit this.
  • the data merging module is also used to transmit the merged data, including: the data merging module is used to transmit the merged data to the central processing unit; the central processing unit is used to transmit the merged data Perform post-processing to obtain the second data of the first radar sensor.
  • the specific implementation of the data merging module for merging data can be divided into data packaging and parallel conversion according to the difference in transmission bit width between the data merging module and the next-level processor (such as the central processing unit).
  • the next-level processor such as the central processing unit.
  • the data merging module is used for the M Combining the first data output by each radar chip includes: the data combining module is used to pack the first data output by each radar chip.
  • the signal transmission bit width between the data merging module and the central processing unit is greater than or equal to the bit width of the combined data of the first radar sensor, that is, the transmission channel between the data merging module and the central processing unit can be complete
  • the combined data of the first radar sensor is transmitted to the ground. Therefore, the data combining module only needs to package the first data output by each of the M radar chips, and then send the combined data to the central processing unit.
  • the data merging module is used for the M radar units.
  • Combining the first data output by each slice includes: the data merging module is used to convert the first data output by the M radar single slices from parallel data into serial data; the data merging module is used to buffer the serial data Data; the data merging module is used to transmit the merged data to the central processing unit, including: the data merging module is used to transmit the serial data to the central processing unit in N times, where N is an integer greater than 1.
  • the data merging module when the signal transmission bit width between the data merging module and the central processing unit is smaller than the bit width of the combined data of the first radar sensor, the data merging module cannot transmit the combined data to the central processing unit at one time through the transmission channel. Therefore, the data merging module needs to convert the merged data from parallel data to serial data and buffer it, and then transmit it to the central processing unit in stages, so that when the transmission bit width is smaller than the merged data bit width, the data merging module can transmit to the central processing unit Complete consolidated data.
  • the radar system includes a forward radar sensor 301, a backward radar sensor 302, and a four-angle radar sensor 303. There are six radar sensors in total, and a central The processor 304, wherein each radar sensor has the same structure, and each radar sensor is a cascaded radar sensor, which respectively includes M radar monoliths and a data merging module. Optionally, in each radar sensor, the data is merged There may be one module or multiple modules, which is not limited in the embodiment of this application; alternatively, the radar chip may be implemented by a system on chip (SoC), and each radar chip is implemented separately It includes a radio frequency unit and a microprocessor.
  • SoC system on chip
  • the radio frequency unit may include a transceiving antenna, for example, a transceiving antenna that receives multiple transmissions.
  • the radar sensor in Figure 4 is a radar sensor formed by cascading four SoC radar chips 401.
  • Each SoC radar chip 401 is provided with a radio frequency front end (RF Front-end) 4011 and a microprocessor 4012.
  • the RF front-end 4011 provides four receiving antennas RX1 to RX4 and two transmitting antennas TX1 and TX2.
  • Each SoC radar monolithic 401 constitutes a structure of 2 transmitters and 4 receivers, and four SoC radars
  • the single chips 401 are cascaded together to form a radar antenna array with 8 transmitters and 16 receivers.
  • the microprocessors 4011 on the four SoC radar chips 401 are connected to the same data merging module 402.
  • the data merging module 402 sends to the central processing unit via Ethernet.
  • the device sends the combined data.
  • each RF front end 4011 may also include a power amplifier (PA), a low noise amplifier (low noise amplifier, LNA), a mixer, an intermediate frequency filter, and other RF circuits, and a chirp parameter configuration register for the chirp signal .
  • PA power amplifier
  • LNA low noise amplifier
  • the cascaded sensor composed of multiple radars cascaded has a relatively large amount of data.
  • it will pose a greater challenge to the data throughput and the computing power of the processor.
  • the solution provided in this embodiment is equipped with a microprocessor on each radar chip of the cascaded radar sensor.
  • the microprocessor preprocesses the echo data obtained by the radio frequency unit, and then the radar is processed by the data merging module.
  • the single-chip pre-processed data of each radar in the sensor is merged to obtain the merged data, and then the merged data is post-processed by the central processing unit, so that the microprocessor and the central processing unit share the computing power and pass the data of the multi-level domain.
  • the processing method avoids the computational burden of the microprocessor or central processing unit caused by the increase in the amount of cascaded radar data. This solves the bottleneck problem of processing capacity, cache capacity and data throughput capacity in large-scale antenna array radar sensors.
  • the central processing unit can be a multi-domain controller (MDC) with strong computing power and rich interfaces. .
  • MDC multi-domain controller
  • the post-processing performed by the above-mentioned central processing unit is not to fuse the processed data of each radar sensor, but to perform further processing on the preprocessed data of the first radar sensor to obtain the first radar sensor.
  • the following describes the detailed flow of pre-processing and post-processing.
  • the first microprocessor is used to preprocess the echo data to obtain the first data, including:
  • the first microprocessor is used to perform ranging processing on the echo data to obtain ranging data.
  • the ranging processing specifically includes:
  • the first microprocessor performs data zero padding on the echo data
  • the first microprocessor performs windowing processing on the echo data after the data is zeroed
  • the first microprocessor performs fast Fourier transform on the windowed echo data to obtain first spectrum data, and the first spectrum data includes at least one first spectrum.
  • the first radio frequency unit may have a multiple-transmit and multiple-receive structure and includes multiple receiving antennas, the echo data received by each receiving antenna constitutes a first frequency spectrum.
  • the first frequency spectrum can be represented by a matrix including horizontal rows and vertical columns, wherein each row records a different detection point detected by echo data, and each column records a different echo data location. The same detection point detected.
  • each single frequency corresponds to a detection point detected by one echo data, so that the ranging data of each detection point can be obtained, and the detection point is a point in the point cloud information.
  • the data merging module is used for merging the first data output by each of the M radar chips to obtain the merged data of the first radar sensor, including:
  • the data merging module is used for merging the ranging data output by each of the M radar chips to obtain the merged data of the first radar sensor.
  • the central processing unit performs post-processing on the combined data to obtain the second data of the first radar sensor, including:
  • the central processor performs speed measurement processing and angle measurement processing on the combined data to obtain the speed measurement data and angle measurement data of the first radar sensor.
  • the specific process of the speed measurement processing includes:
  • the central processing unit obtains the speed data of each detection point according to the distance data corresponding to each single frequency in the first spectrum data.
  • each single frequency in the first spectrum data corresponds to the ranging data of a detection point, that is, distance data
  • the central processing unit when the central processing unit is performing speed measurement operations, since each column in the matrix of the first spectrum represents the same detection Point data under different echo data, so fast Fourier transform is performed on each column to obtain velocity measurement data.
  • a second frequency spectrum is obtained.
  • the second frequency spectrum records the ranging data and velocity measurement of each detection point data.
  • the angle measurement processing can be divided into two working modes: tracking before detection (TBD) and tracking after detection (TAD). The descriptions are made separately below.
  • TBD Tracking before detection
  • the central processing unit obtains second spectrum data.
  • the second spectrum data is spectrum data obtained after the first spectrum data has been processed by distance measurement and speed measurement.
  • the second spectrum data includes at least one second spectrum, and the second spectrum is
  • the first frequency spectrum is the frequency spectrum obtained after ranging and speed measurement processing.
  • the central processor accumulates all the second spectra in the second spectrum data, and the accumulation may be coherent accumulation or incoherent accumulation.
  • the central processing unit performs fast Fourier transform on the detection points at the same position, thereby obtaining the angle measurement data of each detection point.
  • the central processing unit measures the angle of each point separately, so as to obtain the distance, speed and angle data of each point; then the central processing unit screens out the points with stable changes in distance, speed and angle as the detection points and Continuous testing. So as to realize the work of the radar system.
  • the combined data received by the central processing unit is processed by micro-processing for distance measurement. Therefore, the computing power of the central processing unit can be detected first to obtain the trend of the distance and speed of each detection point. Obtain the detection point with linear change in speed and distance from all detection points as the first target detection point, and then track the first target detection point, so that the most sufficient point cloud information can be ensured to improve the accuracy of radar detection.
  • TAD Tracking after detection
  • the central processing unit obtains the second spectrum data
  • the central processor accumulates all the second spectra in the second spectrum data, and the accumulation may be coherent accumulation or incoherent accumulation.
  • the central processing unit performs a two-dimensional constant false alarm rate on the accumulated second spectrum data to obtain a first target detection point from the detection points, and the first target detection point is a part of the detection points;
  • the central processing unit performs fast Fourier transform on each first target detection point respectively, so as to obtain angle measurement data of each first target detection point.
  • the central processing unit performs coherent accumulation or non-coherent accumulation on the first spectrum data in the combined data to perform a two-dimensional constant false alarm ratio (CFAR), by interpreting the 2-dimensional-CFAR
  • the first target detection point is acquired, and then the angle fast Fourier transform is performed on each first target detection point in the second spectrum data to obtain the angle data of the first target detection point.
  • the first target detection point is detected by the 2-dimensional-CFAR method, and then the first target detection point is tracked, so that the computing power of the central processing unit can be saved.
  • the embodiment of the present application can support two working modes of TAD or TBD, and those skilled in the art can choose according to needs, and the embodiments of the present application are not limited.
  • the above-mentioned echo data can be realized by a chirp signal.
  • the specific working mode of the chirp signal in the radar system provided in the embodiment of the present application will be described in detail below.
  • the method further includes:
  • the first radio frequency unit sends a first chirp signal.
  • the first chirp signal is a signal used for detection.
  • the echo data received by the first radio frequency unit is specifically:
  • the first radio frequency unit is used for the second chirp signal.
  • the second chirp signal is a signal that the first chirp signal detects the object to be detected after the first chirp signal is sent by the first radio frequency unit.
  • each radar chip of the M radar chips outputs a second digital chirp signal ranging data, and the ranging data output by the M radar chips are combined to obtain the first radar sensor The data is merged, and when the data amount of the second digital chirp signal corresponding to the ranging data is accumulated to one frame, the speed measurement processing and the angle measurement processing are performed on the second digital chirp signal.
  • second digital chirp signals constitute a frame of second digital chirp signals. How many second digital chirp signals are contained in a specific frame depends on the resolution of the radar system. The more the resolution of the radar system is High, the greater the number of second digital chirp signals contained in one frame, the skilled person can set the number of second digital chirp signals corresponding to one frame of second digital chirp signals according to actual work requirements. For example, the duration of the chirp signal is 15 us. In order to measure the speed, 64 chirp signals are accumulated per frame, so the original echo data size is about 8M bytes. Since the central processing unit cannot determine the speed of the detection point only based on a set of ranging data, the central processing unit needs to wait until the received second chirp signal reaches one frame before starting the speed measurement processing and the angle measurement processing.
  • the radio frequency unit of the radar system transmits the first chirp signal through the transmitting antenna, and then receives the second chirp signal after the echo of the first chirp signal through the receiving antenna.
  • a transmitting antenna transmits 64 first chirp signals in a frame Signal chirps signal
  • each chirps signal detects 1024 detection points
  • the receiving antenna will also receive 64 chirps signals in a frame
  • each chirps signal detects 1024 detection points
  • the chirps signal received by the receiving antenna After the induction, the first frequency spectrum is obtained, and the first frequency spectrum can be expressed as the first two-dimensional matrix shown in the following formula 1:
  • each detection point of each chirps signal constitutes a sample (Sample, S)
  • S 1,1 represents the first chirps signal Detection point
  • S 1,1024 represents the 1024th detection point of the first chirps signal
  • S 64,1024 represents the 1024th detection point of the 64th chirps signal.
  • the first microprocessor is in the process of preprocessing the echo data to obtain the first data.
  • the first microprocessor is in the process of preprocessing the echo data to obtain the first data.
  • a row of the matrix in the above formula 1 can be obtained, for example, the first chirps signal S1 in the second chirps signal, A microprocessor performs fast Fourier transform on S1 in real time to obtain ranging data S 1,1 to S 1,1024 , thereby completing the ranging processing of the first chirps signal in the second chirp signal to obtain the above
  • the first row of ranging data in formula 1 is then combined by the data merging module to send the first row of ranging data to the central processing unit.
  • the central processing unit performs speed measurement processing on the combined data to obtain the speed measurement data of the first radar sensor, including:
  • the central processing unit When the second chirp signal received by the central processing unit is accumulated to one frame, the central processing unit obtains the data of all rows of the first two-dimensional matrix shown in the above formula 1.
  • the central processing unit performs fast Fourier change on the detection points of each column of the first two-dimensional matrix, thereby obtaining the speed information of each detection point, thereby completing the speed measurement process, for example, for the row S 1,1 to S 64,1 Perform fast Fourier transformation to obtain RD 1,1 to RD 64,1 , thereby obtaining a second frequency spectrum, which can be expressed as the following second two-dimensional matrix:
  • each detection point S undergoes two fast Fourier transforms to obtain a detection point RD, where each detection point RD is a single frequency, and each single frequency
  • the distance measurement data and speed measurement data of the detection point are recorded.
  • RD 1,1 indicates that the distance of the detection point is 3m, and the speed is 2m/s
  • RD 1, 2 indicates that the distance of the detection point is 4m, and the speed is 5m. /s
  • RD 1,1024 means that the distance of the detection point is 1024m, and the speed is 5m/s.
  • the radio frequency unit may have a multi-transmit and multi-receive structure, including multiple receiving antennas. Therefore, the echo data received by each receiving antenna constitutes a first frequency spectrum.
  • multiple second frequency spectra are obtained. As shown in Figure 11, taking a radio frequency unit with 2 transmitters and 4 receivers as an example, since there are 4 receiving antennas, each second chirp signal will be The receiving antenna receives the four second frequency spectra from RD-map1 to RD-map4 accordingly.
  • the angle measurement processing of the second spectrum data can be divided into two working modes, TAD or TBD, which will be specifically described below.
  • the central processing unit accumulates all four second frequency spectra to obtain the second frequency spectrum data as shown in FIG. 11.
  • the central processing unit performs fast Fourier transform on the detection points at the same position, thereby obtaining the angle measurement data of each detection point.
  • the detection points RD 1,1 on each second spectrum RD-map record the first chirps signal in the second chirp signal
  • the central processor screens out the detection points with stable changes in distance, speed, and angle for continuous, so as to realize the work of the radar system.
  • the central processor performs a two-dimensional constant false alarm ratio (CFAR) on all the second spectrums in the second spectrum data to obtain the first target detection point, which is the distance and speed of the first target detection point.
  • CFAR constant false alarm ratio
  • the central processing unit accumulates all the second frequency spectrums in the second frequency spectrum data.
  • the accumulation may be coherent accumulation or non-coherent accumulation.
  • the second spectrum data as shown in FIG. 11 is obtained.
  • the central processing unit performs fast Fourier transform on each first target detection point respectively, so as to obtain angle measurement data of each first target detection point.
  • each RD-map in RD-map1 to RD-map4 contains the information of all detection points; while in the working mode of TAD, each RD-map1 to RD-map4 Each RD-map contains the information of the first target detection point.
  • each RD-map contains one RD 1,1 for four RDs.
  • -map corresponds to the four RD 1,1 to perform fast Fourier change, so as to obtain the angle information of RD 1,1 , and finally obtain the distance, speed and angle of the first first target detection point.
  • the same processing is performed on each first target detection point on the RD-map, and the distance, speed and angle of all the first target detection points can be obtained. So as to realize the work of the radar system.
  • the microprocessor compresses the preprocessed data, and the central processing unit decompresses the received combined data, thereby reducing the data
  • the amount of data transmitted improves data throughput.
  • the specific process of compression and decompression is described in detail below.
  • the above-mentioned first microprocessor preprocesses the echo data to obtain the first data, including:
  • the first microprocessor compresses the echo data to obtain the first compressed data.
  • the first microprocessor performs the compression only after processing the echo data.
  • the processing of the echo data by the first microprocessor can be any of the above-mentioned processing methods, see The foregoing description is to be understood, and will not be repeated here.
  • the aforementioned data merging module is used to merge the first data output by each of the M radar chips to obtain the merged data of the first radar sensor, including:
  • the data merging module is used for merging the first compressed data output by each of the M radar chips to obtain the compressed and merged data of the first radar sensor.
  • the data merging module merges the compressed data to obtain compressed merged data.
  • the central processing unit is used for post-processing the combined data, including:
  • the central processing unit is used to decompress the compressed and combined data.
  • the central processing unit decompresses it to perform subsequent steps.
  • the technical solution used in the above compression and decompression step can be any compression and decompression solution.
  • the data compression algorithm can rely on the result information of the distance fast Fourier change or the time domain information of the original echo data to achieve lossy compression.
  • lossless compression which is not limited in the embodiment of the present application.
  • the first microprocessor before the first microprocessor starts processing the echo data, it can also down-sample the echo data, thereby further compressing the amount of data to be processed.
  • the specific amount of sampled data to be down-sampled can be determined according to actual usage requirements. In the following, taking the echo data as the second digital chirp signal as an example, the working mode of down-sampling will be described in detail.
  • the above-mentioned first microprocessor preprocesses the echo data to obtain the first data, including:
  • the first microprocessor down-samples the X second digital chirp signals received by the first radio frequency unit to obtain Y third digital chirp signals, where X and Y are both positive integers greater than 1, where, Y is less than X.
  • the first microprocessor only performs sampling processing on a part of it, thereby reducing the amount of data that needs to be processed.
  • the data merging module is used to merge the first data output by each of the M radar monoliths to obtain the merged data of the first radar sensor, including:
  • the data merging module is used to merge the third digital chirp signals output by each of the M radar chips to obtain the merged data of the first radar sensor.
  • the third digital chirp signal output by each of the M radar chips is the third digital chirp signal after preprocessing.
  • the central processor is used for post-processing the combined data to obtain the second data of the first radar sensor, including:
  • the central processing unit is used for post-processing the third digital chirp signal in the combined data to obtain the second data of the first radar sensor.
  • Micro-processing down-sampling-microprocessor ranging-data merging module merged data-central processing unit speed and angle measurement.
  • the microprocessor down-samples the chirp signal received by the radio frequency unit, and then the microprocessor performs a ranging operation on the down-sampled data. After being combined and sent by the data combining unit, the central processing unit performs a range measurement on the combined data. Perform speed and angle measurement operations.
  • the down-sampling ranging, velocity measurement and angle measurement operations, reference may be made to the above-mentioned record, which will not be repeated here.
  • Micro-processing down sampling-microprocessor ranging-microprocessor compressed data-data merging module merged data-central processing unit decompression, speed measurement, angle measurement.
  • the microprocessor down-samples the chirp signal received by the radio frequency unit, and then the microprocessor performs ranging processing on the down-sampled data, and then the microprocessor compresses the ranging processed data. After the compressed data is combined and sent by the data combining unit, the central processing unit performs decompression, velocity measurement and angle measurement operations on the combined data.
  • the central processing unit performs decompression, velocity measurement and angle measurement operations on the combined data.
  • Micro-processing down-sampling-data merging module merging data-central processing unit ranging, speed and angle measurement.
  • the microprocessor only down-samples the chirp signal, and then sends the down-sampled data to the central processing unit through the data merging module for distance measurement, speed measurement, and angle measurement.
  • the sampling step the amount of data is reduced, so the problem of data throughput bottleneck can also be solved.
  • the specific implementation of the above-mentioned down-sampling, ranging, velocity and angle measurement processing can be referred to the above-mentioned record, and will not be repeated here.
  • Micro-processing down-sampling-micro-processing compressed data-data merging module merging data-central processing unit decompression, distance measurement, speed measurement, angle measurement.
  • the microprocessor only down-samples the chirp signal, and compresses the down-sampled data, and then sends the compressed data to the central processing unit through the data merging module for decompression ranging, speed measurement, and measurement.
  • Angle processing because the chirp signal has undergone down-sampling, the amount of data is reduced.
  • the data after down-sampling is compressed, which further solves the problem of data throughput bottlenecks.
  • the embodiment of the present application also provides another radar system.
  • the difference between this radar system and the above-mentioned radar system lies in the radar sensor.
  • Each radar sensor includes a microprocessor and at least one radio frequency unit.
  • the radar sensor can be implemented in the following two ways: :
  • the radar sensor includes a microprocessor 501 and four radar chips 502, wherein all the four radar chips 502 are connected to the same microprocessor 501, and the micro processor
  • the device 501 processes the echo data obtained by the four radar chips 502.
  • the radar sensor may also be at least one radar single chip 502 and a microprocessor group.
  • the microprocessor group includes at least one microprocessor 501, and the microprocessor group is used to control at least one radar single chip. 502 echo data is processed.
  • the radar chip 502 includes a radio frequency unit, a power amplifier (PA), a low noise amplifier (LNA), a mixer, an intermediate frequency filter IF, and other RF circuits, and chirp parameter configuration of a chirp signal Register, where the radio frequency unit may include transceiver antennas, for example, it may include four receiving antennas RX1 to RX4 and two transmitting antennas TX1 and TX2.
  • the radio frequency unit may include transceiver antennas, for example, it may include four receiving antennas RX1 to RX4 and two transmitting antennas TX1 and TX2.
  • Each radar chip 502 constitutes a 2-transmitting 4-receiving structure, and four radar monoliths 502 is cascaded together to form a radar antenna array with 8 transmitters and 16 receivers.
  • RX1 to RX4 are respectively connected to an LNA, and each LNA is connected to IF and ADC through its own mixer.
  • the ADC of each radar chip 502 and The same microprocessor 501
  • the radar sensor includes a microprocessor and a radio frequency unit connected to the microprocessor, wherein the radio frequency unit includes a separate receiving end radio frequency front end (RX RF front end) and A transmitting radio frequency front end (TX RF front end), where the RX RF front end is connected to the microprocessor through ADC and sends echo data to the microprocessor, and the TX RF front end is connected to the microprocessor through the voltage controlled oscillator and Chirp parameter configurator The device is connected and receives the signal sent by the microprocessor.
  • RX RF front end receives the signal sent by the microprocessor.
  • TX RF front end radio frequency front end
  • TX RF front end transmitting radio frequency front end
  • the device is connected and receives the signal sent by the microprocessor.
  • the difference between another radar system provided by this application and the aforementioned radar system is that because the radio frequency unit is connected to the microprocessor in a many-to-one or one-to-one manner, it is no longer needed
  • the data merging unit merges the echo data obtained by each radio frequency unit, but the microprocessor directly receives the echo data sent by each radio frequency unit, and the microprocessor sends the preprocessed echo data To the central processor.
  • the same radar is arranged around the car body.
  • the vehicle-mounted radar system provided by the first specific embodiment of the present application includes a vehicle body 700, a central processing unit 704, and forward radar sensors 701, 701 and 701 respectively arranged on the front of the vehicle body.
  • the rear-facing radar sensor 702 and the four-corner angle radar sensor 703 on the back are six radar sensors; the forward radar sensor 701, the backward radar sensor 702, and the angle radar sensor 703 are respectively connected to the central processing unit 704, wherein
  • the above-mentioned six radar sensors have the same structure, which can be the radar sensor described in any of the above-mentioned radar systems.
  • central processing unit 704 may be a central radar-signal processing unit (central RPU).
  • central RPU central radar-signal processing unit
  • the radar sensor provided in the embodiment of the application is arranged around the vehicle body, and the radar sensor is connected to the central processing unit to realize the multi-level domain data method provided in the embodiment of the application, so that the radar
  • the microprocessor and central processing unit in the sensor share the computing power pressure of data processing, which improves the processing capacity, cache capacity and data throughput capacity of the radar system to adapt to the huge amount of data brought by the large-scale antenna array radar sensor .
  • the forward and backward radar sensors in the radar system require a larger detection range and higher point cloud density, while the diagonal radar only needs a medium detection range and point cloud. density. Therefore, it is possible to perform differentiated processing for radar sensors with different detection azimuths in the radar system, and a specific description will be given below for this situation.
  • Different radars are arranged around the car body.
  • the vehicle-mounted radar system provided by the second specific embodiment of the present application includes a vehicle body 800, a central processing unit 804, and forward radar sensors 801,
  • the forward radar sensor 801 and the backward radar sensor 802 are radar sensors provided in the embodiments of the application.
  • the radar may specifically include a single radar chip, which includes a radio frequency unit, a power amplifier (PA), a low noise amplifier (LNA), a mixer, RF circuits such as intermediate frequency filter IF and chirp parameter configuration registers for chirp signals.
  • PA power amplifier
  • LNA low noise amplifier
  • RF circuits such as intermediate frequency filter IF and chirp parameter configuration registers for chirp signals.
  • the radio frequency unit may include a transceiver antenna, for example, it may include four receiving antennas RX1 to RX4 and two transmitting antennas TX1 and TX2, 2 transmitting 4 receiving Structure, where each receiving antenna in RX1 to RX4 is connected to an LNA respectively, and each LNA is connected to IF and ADC in turn through its own mixer, and finally connected to the same microprocessor. Since the angular radar sensor only needs a medium detection range and point cloud density, the amount of data is small.
  • the microprocessor can independently process the echo data of the radar single chip, including range measurement processing, speed measurement processing and angle measurement processing. The processed data is sent to the central processing unit 804.
  • the forward radar sensor and the backward radar sensor use the radar sensor provided in the embodiment of the present application, and the angle radar uses the radar sensor in the prior art. Please refer to FIG. 10.
  • the processing of data by the central processing unit includes the following working steps.
  • the central processing unit receives data from the radar sensor.
  • the data may be merged data sent by the data merging module of the radar sensor, or may be data sent directly by the microprocessor of the radar sensor.
  • the data merging module of the radar sensor may be data sent directly by the microprocessor of the radar sensor.
  • the central processing unit can receive data from radar sensors via Ethernet,
  • the central processor determines whether the received data comes from a forward radar sensor or a backward radar sensor.
  • the radar sensor when the radar sensor sends data to the central processing unit, it will tag the sent data, so that the central processing unit can determine the radar sensor sending the data based on the label.
  • steps 1003 to 1004 are executed.
  • the central processing unit obtains the point cloud data in the data.
  • the data received by the central processing unit is data processed by the angular radar sensor, so the central processing unit can directly obtain the point cloud data.
  • the central processing unit implements moving target tracking according to the point cloud data.
  • the central processing unit implements moving target tracking according to the point cloud data to complete the tracking of the scanned target.
  • step 1005 is executed.
  • the central processing unit processes the data.
  • the processing of data by the central processing unit can be any of the processing methods of the central processing unit disclosed in the radar system provided in this embodiment of the application, which can be understood with the above-mentioned parameters, and will not be repeated here. .
  • the central processing unit uses the processing procedure provided in the embodiment of this application to perform subsequent processing.
  • the central processing unit can obtain point cloud data and perform subsequent moving target tracking processing.
  • the central processing unit After performing differentiated processing on different radar sensor data, the central processing unit performs 360-degree radar data fusion and heterogeneous sensor data fusion. In this way, different radar sensors arranged around the car body can be processed according to actual needs, which can adapt to different usage requirements and save computing power.
  • the disclosed system can be implemented in other ways.
  • the device embodiments described above are only illustrative.
  • the division of the units is only a logical function division, and there may be other divisions in actual implementation, for example, multiple units or components can be combined or It can be integrated into another system, or some features can be ignored or not implemented.
  • the displayed or discussed mutual coupling or direct coupling or communication connection may be indirect coupling or communication connection through some interfaces, devices or units, and may be in electrical, mechanical or other forms.
  • the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the objectives of the solutions of the embodiments.
  • each unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist alone physically, or two or more units may be integrated into one unit.
  • the above-mentioned integrated unit can be implemented in the form of hardware or software functional unit.
  • the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer readable storage medium.
  • the technical solution of this application essentially or the part that contributes to the existing technology or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium , Including several instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute all or part of the steps of the method described in each embodiment of the present application.
  • the aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disk or optical disk and other media that can store program code .

Landscapes

  • Engineering & Computer Science (AREA)
  • Radar, Positioning & Navigation (AREA)
  • Remote Sensing (AREA)
  • Physics & Mathematics (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • General Physics & Mathematics (AREA)
  • Electromagnetism (AREA)
  • Radar Systems Or Details Thereof (AREA)

Abstract

一种雷达系统，可应用于自动驾驶的车载雷达，例如车载毫米波雷达；包括至少一个雷达传感器，至少一个雷达传感器中的第一雷达传感器包括数据合并模块(402)和多个雷达单片(401)，其中，每个雷达单片(401)包括第一射频前端(4011)和第一微处理器(4012)；第一微处理器(4012)用于对第一射频前端(4011)获取的回波数据进行预处理，经过数据合并模块(402)的合并和传输后，由下一级别的处理器对预处理后的回波数据进行后处理，从而生成雷达系统的点云数据。通过第一微处理器(4012)预处理回波数据；减少了后续处理器的算力负担，提升了微处理器的缓存能力和数据吞吐能力，资源分配更合理，从而提升了雷达系统对大规模天线阵列雷达传感器的适应性。

Description

一种雷达系统 技术领域
本申请涉及雷达技术领域，具体涉及一种雷达系统。
背景技术
高分辨率车载雷达是自动驾驶中不可缺少的重要组成部分，用于测量目标距离和速度等。在自动驾驶领域，雷达传感器需要提供更强的环境感知能力。增加雷达传感器阵列天线规模是一种典型的提高传感器对环境测量能力的方法。在测速和测距的基础上，更多的收发天线可以增强雷达传感器对目标角度的分辨能力。相应的，大规模天线阵列带来更强的数据吞吐需求、更多的数据存储需求以及更大的算力需求。因此，后续的信号处理和数据处理能力也需要匹配逐步提升的传感器规格。
然而，车身可以安装雷达的位置以及相应的安装尺寸并没有持续增大，随着雷达传感器的天线阵列规模增大，接收天线数量大幅增加，现有技术中，对于多个雷达传感器的数据处理，若采用中央处理器统一处理的方式，将会给中央处理器的算力带来巨大挑战；若采用雷达传感器一端的处理器处理后发送给中央处理器进行汇总的方式，雷达传感器一端的处理器会很快进入到处理能力、缓存能力和数据吞吐能力的瓶颈；这些缺点将制约雷达系统对大规模天线阵列雷达传感器的适应性。
因此，现有技术中存在的上述问题还有待于改善。
发明内容
本发明实施例提供了一种雷达系统，能够通过微处理器对雷达传感器的获取的数据进行预处理，减少后续处理器的处理压力，从而提升雷达系统的数据处理能力。
本申请第一方面提供一种雷达系统，包括至少一个雷达传感器，该至少一个雷达传感器中的第一雷达传感器包括数据合并模块和M个雷达单片，其中，每个雷达单片包括第一射频单元和第一微处理器，该M为大于1的整数；其中，该第一微处理器可以是雷达信号处理单元(radar-signal processing unit，RPU),该雷达单片可以是由分离模块组成的电路结构，也可以是片上系统(system on chip，SoC)；该第一射频单元用于接收回波数据，其中，该回波数据可以是该第一射频单元发射线性调频信号(chirp)的回波数据；该第一微处理器用于对该回波数据进行预处理，以得到第一数据；该数据合并模块用于对该M个雷达单片各自输出的第一数据进行合并，以得到该第一雷达传感器的合并数据；该数据合并模块还用于传输该合并数据，该合并数据用于经后处理后得到该第一雷达传感器的第二数据，该至少一个雷达传感器各自的第二数据用于生成该雷达系统的点云信息。
本实施例中，第一雷达传感器中每个雷达单片分别包含第一射频单元和第一微处理器，在每个雷达单片中，第一微处理器对第一射频单元所获取的回波数据进行预处理后得到第一数据，并将该第一数据发送给数据合并模块，该数据合并模块与所有雷达单片连接，数据合并模块在接收到各个雷达单片发来的第一数据后，将所有第一数据合并为该合并数据，该合并数据即为该第一雷达传感器经过预处理后的数据，进一步地，后续只需要对合并数据进行进一步处理，即可获得该第一雷达传感器的第二数据，雷达系统中包含多个雷达传 感器，所有雷达传感器的第二数据最终用于生成雷达系统的点云数据。上述工作过程中，对于所获取的第一射频单元所获取的回波数据，并不是完全由雷达传感器的微处理器处理，而是由雷达传感器中的微处理器进行预处理后，发送到下一个环节进行后处理，这样一来，对于回波数据处理的算力压力，由雷达传感器和下一级别的设备共同分担，从而解决了雷达传感器的天线阵列规模增大时，雷达传感器一端的处理器会很快进入到处理能力、缓存能力和数据吞吐能力的瓶颈问题，提升了雷达系统对大规模天线阵列雷达传感器的适应性。
结合上述第一方面，在第一种可能的实现方式中，该雷达系统还包括中央处理器，该该中央处理器可以是中央雷达信号处理单元(central radar-signal processing unit,central RPU)该数据合并模块还用于传输该合并数据，包括：该数据合并模块用于向该中央处理器传输该合并数据；该中央处理器用于对该合并数据进行后处理，以得到该第一雷达传感器的第二数据。
本实施例中，数据合并模块将合并数据传输给中央处理器，并由中央处理器来对该合并数据进行预处理，以得到第一雷达传感器的第二数据，由于该合并数据已经经过第一微处理器的预处理，因此，中央处理器只需较少算力，即可计算得到第二数据，从而减少了中央处理器的处理压力，同时，由于所传输的合并数据经过预处理，也提升了数据合并模块和中央处理器之间的数据吞吐能力。
结合上述第一方面第一种可能的实现方式，在第二种可能的实现方式中，该数据合并模块与该中央处理器之间的信号传输位宽大于或等于该第一雷达传感器的合并数据的位宽，则，该数据合并模块用于对该M个雷达单片各自输出的第一数据进行合并，包括：该数据合并模块用于对该M个雷达单片各自输出的第一数据进行打包。
本实施例中，由于数据合并模块与中央处理器之间的信号传输位宽大于或等于第一雷达传感器的合并数据的位宽，即，数据合并模块与中央处理器之间的传输信道能够完整地传输第一雷达传感器的合并数据，因此，数据合并模块只需要对M个雷达单片各自输出的第一数据进行打包，即可将所得到的合并数据发送给中央处理器。
结合上述第一方面第一种可能的实现方式，在第三种可能的实现方式中，该数据合并模块与该中央处理器之间的信号传输位宽小于该第一雷达传感器的合并数据的位宽，则，该数据合并模块用于对该M个雷达单片各自输出的第一数据进行合并，包括：该数据合并模块用于将该M个雷达单片各自输出的第一数据由并行数据转化为串行数据；该数据合并模块用于缓存该串行数据；该数据合并模块用于向该中央处理器传输该合并数据，包括：该数据合并模块用于分N次向该中央处理器传输该串行数据，该N为大于1的整数。
本实施例中，当数据合并模块与该中央处理器之间的信号传输位宽小于第一雷达传感器的合并数据的位宽时，数据合并模块无法通过传输通道一次性向中央处理器传输合并数据，因此，数据合并模块需要将合并数据由并行数据转化为串行数据并缓存，之后分次向中央处理器传输，从而使得传输位宽小于合并数据位宽时，数据合并模块能够向中央处理器传输完整的合并数据。
结合上述第一方面第一至第三种可能的实现方式，在第种可能的实现方式中，该第一微处理器用于对该回波数据进行预处理，以得到第一数据，包括：该第一微处理器用于对 该回波数据进行测距处理，以得到测距数据；该数据合并模块用于对该M个雷达单片各自输出的第一数据进行合并，以得到该第一雷达传感器的合并数据，包括：该数据合并模块用于对该M个雷达单片各自输出的测距数据进行合并，以得到该第一雷达传感器的合并数据；该中央处理器用于对该合并数据进行后处理，以得到该第一雷达传感器的第二数据，包括：该中央处理器用于对该合并数据进行测速处理和测角处理，以得到该第一雷达传感器的测速数据和测角数据。
本实施例中，雷达传感器一端的第一微处理器用于对回波数据进行测距处理，以完成预处理的操作，之后由中央处理器来对经过测距处理的数据进行测速处理和测角处理，从而使得第一微处理器和中央处理器分担了对回波数据的算力压力，提升了数据的处理能力，同时，由于回波数据已经经过测距处理，数据量变小，从而提升了数据吞吐能力。
结合上述第一方面第四种可能的实现方式，在第五种可能的实现方式中，该第一微处理器用于对该回波数据进行测距处理，以得到测距数据，包括：该第一微处理器用于对该回波数据进行数据补零；该第一微处理器还用于对数据补零后的该回波数据进行加窗处理；该第一微处理器还用于对加窗后的该回波数据进行快速傅里叶变换，以得到第一频谱数据，该第一频谱数据中包含至少一个第一频谱，其中，该第一频谱中的每个单频对应一个检测点的距离数据，该检测点为该点云信息中的一个点。
本实施例中，第一微处理器对回波数据进行补零以使得回波数据满足处理需求，对补零后的回波数据进行加窗处理后进行快速速傅里叶变换，以得到第一频谱数据，通过对该第一频谱数据进行解读，其中，第一频谱中的每个单频对应一个检测点的距离数据，从而得到了该雷达单片所有检测点的距离信息，后续中央处理器处理的过程中，所有雷达传感器的检测点构成雷达系统的点云。
结合上述第一方面第五种可能的实现方式，在第六种可能的实现方式中，该中央处理器用于对该合并数据进行测速处理和测角处理，以得到该第一雷达传感器的测速数据和测角数据，包括：该中央处理器用于根据该第一频谱数据中每个单频所对应的距离数据，获取各个检测点的测速数据。
本实施例中，由于无法通过单个检测点的距离计算该检测点的速度，因此，需要获得多个距离数据之后，再对各个点进行测速。
结合上述第一方面第六种可能的实现方式，在第七种可能的实现方式中，该中央处理器用于对该合并数据进行测速处理和测角处理，以得到该第一雷达传感器的测速数据和测角数据，包括：
该中央处理器用于获取第二频谱数据，该第二频谱数据为该第一频谱数据经过测距处理和测速处理后得到的频谱数据，其中，该第二频谱数据中包含至少一个第二频谱，该第二频谱为该第一频谱经过测距处理和测速处理后所得到的频谱；
该中央处理器用于对该第二频谱数据中的所有第二频谱进行累加，其中，该累加可以是相干累加，也可以是非相干累加；
在该第二频谱数据的所有该第二频谱中，该中央处理器用于对每个第二频谱同一位置的检测点进行快速傅里叶变换，从而得到每个检测点的测角数据。
本实施例中，第一微处理器对回波数据进行测距处理后得到第一频谱数据，中央处理器对第一频谱数据进行测速处理得到第二频谱数据，之后中央处理器对第二频谱数据中的所有第二频谱进行累加，对第二频谱同一位置的检测点进行快速傅里叶变换，从而得到每个检测点的测角数据；此种方式对所有检测点进行测距、测速及测角，实现对所有检测点的跟踪，之后再从跟踪的检测点中筛选出需要检测的目标，计算量较大，但本实施例中第一微处理器与中央处理器协同工作的方式能够支撑该种工作方式所需要的算力。
结合上述第一方面第六种可能的实现方式，在第八种可能的实现方式中，该中央处理器用于对该合并数据进行测速处理和测角处理，以得到该第一雷达传感器的测速数据和测角数据，包括：该中央处理器用于获取第二频谱数据，该第二频谱数据为该第一频谱数据经过测距处理和测速处理后得到的频谱数据，其中，该第二频谱数据中包含至少一个第二频谱，该第二频谱为该第一频谱经过测距处理和测速处理后所得到的频谱；该中央处理器用于对该第二频谱数据中的所有该第二频谱进行累加，其中，该累加可以是相干累加，也可以是非相干累加；该中央处理器用于对累加后的该第二频谱数据进行二维恒虚警率，以从该检测点中获取第一目标检测点，该第一目标检测点为该检测点中的部分检测点；在该第二频谱数据的所有该第二频谱中，该中央处理器用于分别对每个第一目标检测点进行快速傅里叶变换，从而得到每个第一目标检测点的测角数据。
本实施例中，中央处理器对第一频谱数据进行测速处理后得到第二频谱数据，对第二频谱数据中的所有第二频谱进行累加，之后对累加后的第二频谱数据进行二维恒虚警率，以提取检测点中需要跟踪的部分检测点作为第一目标检测点，之后对每个第一目标检测点进行快速傅里叶变换，从而得到每个第一目标检测点的测角数据，此种方式先检测，之后通过二维恒虚警率筛选出需要跟踪的点进行跟踪，较为节省算力。
结合上述第一方面第一至第三种种可能的实现方式，在第九种可能的实现方式中，该第一射频单元还用于发送第一线性调频信号，该第一线性调频信号用于进行探测；该第一射频单元用于接收回波数据，包括：该第一射频单元用于接收第二线性调频信号，该第二线性调频信号为该第一射频单元发送该第一线性调频信号之后返回的线性调频信号。
本实施例中，第一射频单元发射的第一线性调频信号在触碰到障碍物之后反弹，从而使得第一射频单元能够接收到第二线性调频信号，从而能够进行探测。
结合上述第一方面第九种可能的实现方式，在第十种可能的实现方式中，该第一微处理器用于对该回波数据进行预处理，以得到第一数据，包括：该第一微处理器用于对该第二线性调频信号进行预处理，以得到测距数据；该数据合并模块用于对该M个雷达单片各自输出的测距数据进行合并，以得到该第一雷达传感器的合并数据，包括：该数据合并模块用于，该M个雷达单片中的每个雷达单片每输出一个第二数字线性调频信号的测距数据，对M个雷达单片各自输出的测距数据进行合并，以得到该第一雷达传感器的合并数据；该中央处理器用于对该合并数据进行后处理，以得到该第一雷达传感器的第二数据，包括：
该中央处理器用于，当该测距数据所对应的第二数字线性调频信号的数据量累积达到一帧时，对该第二数字线性调频信号进行测速处理和测角处理。
本实施例中，第一射频单元每接收到一个第二数字线性调频信号，第一微处理器对该 第二数字线性调频信号执行一次测距处理，之后数据合并模块实时合并各个雷达传感器的测距数据并发送给中央处理器；由于需要多组测距数据才能进行测速，因此，当测距数据所对应的第二数字线性调频信号的数据量累积达到一帧时，中央处理器对第二数字线性调频信号进行测速处理和测角处理。
结合上述第一方面第九种可能的实现方式，在第十一种可能的实现方式中，该第一微处理器用于对该回波数据进行预处理，以得到第一数据，包括：
该第一微处理器用于对该第一射频单元所接收的X个该第二数字线性调频信号进行下采样，得到Y个第三数字线性调频信号，其中，X和Y均为大于1的正整数，其中，Y小于X；
该数据合并模块用于对该M个雷达单片各自输出的第一数据进行合并，以得到该第一雷达传感器的合并数据，包括：
该数据合并模块用于对该M个雷达单片各自输出的第三数字线性调频信号进行合并，以得到该第一雷达传感器的合并数据；
该中央处理器用于对该合并数据进后处理，以得到该第一雷达传感器的第二数据，包括：
该中央处理器用于对该合并数据中的第三数字线性调频信号进行后处理，以得到该第一雷达传感器的第二数据。
本实施例中，当第一射频单元的采样量大于雷达系统所需的数量时，第一微处理器需要对第一射频单元所接收的第二数字线性调频信号进行下采样，从而压缩第一射频单元所接收的数据量，以满足雷达系统的工作要求。
结合上述第一方面及第一方面第一至第十一种可能的实现方式，在第十二种可能的实现方式中，该第一微处理器用于对该回波数据进行预处理，以得到第一数据，包括；
该第一微处理器用于对该回波数据进行压缩，以得到第一压缩数据；
该数据合并模块用于对该M个雷达单片各自输出的第一数据进行合并，以得到该第一雷达传感器的合并数据，包括：
该数据合并模块用于对该M个雷达单片各自输出的该第一压缩数据进行合并，以得到该第一雷达传感器的压缩合并数据；
该中央处理器用于对该合并数据进行后处理，包括：
该中央处理器用于对该压缩合并数据进行解压缩。
本实施例中，在第一微处理器对回波数据进行测距处理后，进一步对测距数据进行压缩，从而使得数据合并模块合并并传输压缩后的数据，从而减少了数据合并模块和中央处理器之间的数据传输压力，提升了数据吞吐能力。
从以上技术方案可以看出，本申请实施例具有以下优点：
本发明实施例中，提供了一种雷达系统，可应用于自动驾驶的车载雷达，例如车载毫米波雷达，包括至少一个雷达传感器，该至少一个雷达传感器中的第一雷达传感器包括数据合并模块和M个雷达单片，其中，每个雷达单片包括第一射频单元和第一微处理器，该M为大于1的整数；该第一射频单元用于接收回波数据；该第一微处理器用于对该回波数 据进行预处理，以得到第一数据；该数据合并模块用于对该M个雷达单片各自输出的第一数据进行合并，以得到该第一雷达传感器的合并数据；该数据合并模块还用于传输该合并数据，该合并数据用于经后处理后得到该第一雷达传感器的第二数据，该至少一个雷达传感器各自的第二数据用于生成该雷达系统的点云信息。本实施例中，第一微处理器与中央处理器通过任务分工，使系统能够兼容更大规模天线阵列的雷达传感器、能够适配不同的系统参数和算法处理流程；微处理器和中央处理器的算力负担、硬件资源负担都相应减小，资源分配更合理。同时，由于微处理器和中央处理器共同工作，提升了处理效率。
本申请第二方面提供一种雷达系统，包括至少一个雷达传感器，该至少一个雷达传感器中的第一雷达传感器包括第一微处理器和至少一个第一射频单元；其中，该第一微处理器可以是雷达信号处理单元(radar-signal processing unit，RPU)；该至少一个第一射频单元用于接收至少一个回波数据，其中，该回波数据可以是该第一射频单元发射线性调频信号(chirp)的回波数据；该第一微处理器用于对该至少一个回波数据进行预处理，以得到第一数据；该第一微处理器还用于传输该第一数据，该第一数据用于经后处理后得到该第一雷达传感器的第二数据，该至少一个雷达传感器各自的第二数据用于生成该雷达系统的点云信息。
本实施例中，第一雷达传感器包含第一微处理器和至少一个第一射频单元，第一微处理器对第一射频单元所获取的回波数据进行预处理后得到第一数据，并将该第一数据传输给下一级别的处理器，后续只需要对第一数据进行进一步处理，即可获得该第一雷达传感器的第二数据，雷达系统中包含多个雷达传感器，所有雷达传感器的第二数据最终用于生成雷达系统的点云数据。上述工作过程中，对于所获取的第一射频单元所获取的回波数据，并不是完全由雷达传感器的第一微处理器处理，而是由雷达传感器中的第一微处理器进行预处理后，发送到下一个环节进行后处理，这样一来，对于回波数据处理的算力压力，由雷达传感器和下一级别的设备共同分担，从而解决了雷达传感器的天线阵列规模增大时，雷达传感器一端的处理器会很快进入到处理能力、缓存能力和数据吞吐能力的瓶颈问题，提升了雷达系统对大规模天线阵列雷达传感器的适应性。
结合上述第二方面，在第一种可能的实现方式中，该雷达系统还包括中央处理器，该该中央处理器可以是中央雷达信号处理单元(central radar-signal processing unit,central RPU)，该第一微处理器还用于传输该第一数据，包括：该第一微处理器用于向该中央处理器传输该第一数据；该中央处理器用于对该第一数据进行后处理，以得到该第一雷达传感器的第二数据。
本实施例中，第一微处理器将第一数据传输给中央处理器，并由中央处理器来对该第一数据进行预处理，以得到第一雷达传感器的第二数据，由于该第一数据已经经过第一微处理器的预处理，因此，中央处理器只需较少算力，即可计算得到第二数据，从而减少了中央处理器的处理压力，同时，由于所传输的第一数据经过预处理，也提升了第一微处理器和中央处理器之间的数据吞吐能力。
结合上述第二方面第一种可能的实现方式，在第二种可能的实现方式中，该第一微处理器用于对该回波数据进行预处理，以得到第一数据，包括：该第一微处理器用于对该回 波数据进行测距处理，以得到测距数据；该中央处理器用于对该第一数据进行后处理，以得到该第一雷达传感器的第二数据，包括：该中央处理器用于对该第一数据进行测速处理和测角处理，以得到该第一雷达传感器的测速数据和测角数据。
本实施例中，第一微处理器用于对回波数据进行测距处理，以完成预处理的操作，之后由中央处理器来对经过测距处理的数据进行测速处理和测角处理，从而使得第一微处理器和中央处理器分担了对回波数据的算力压力，提升了数据的处理能力，同时，由于回波数据已经经过测距处理，数据量变小，从而提升了数据吞吐能力。
结合上述第二方面第二种可能的实现方式，在第三种可能的实现方式中，该第一微处理器用于对该回波数据进行测距处理，以得到测距数据，包括：该第一微处理器用于对该回波数据进行数据补零；该第一微处理器还用于对数据补零后的该回波数据进行加窗处理；该第一微处理器还用于对加窗后的该回波数据进行快速傅里叶变换，以得到第一频谱数据，该第一频谱数据中包含至少一个第一频谱，其中，该第一频谱中的每个单频对应一个检测点的距离数据，该检测点为该点云信息中的一个点。
本实施例中，第一微处理器对回波数据进行补零以使得回波数据满足处理需求，对补零后的回波数据进行加窗处理后进行快速速傅里叶变换，以得到第一频谱数据，通过对该第一频谱数据进行解读，其中，第一频谱中的每个单频对应一个检测点的距离数据，从而得到了该雷达单片所有检测点的距离信息，后续中央处理器处理的过程中，所有雷达传感器的检测点构成雷达系统的点云。
结合上述第二方面第三种可能的实现方式，在第四种可能的实现方式中，该中央处理器用于对该第一数据进行测速处理和测角处理，以得到该第一雷达传感器的测速数据和测角数据，包括：该中央处理器用于根据该第一频谱数据中每个单频所对应的距离数据，获取各个检测点的测速数据。
本实施例中，由于无法通过单个检测点的距离计算该检测点的速度，因此，需要获得多个距离数据之后，再对各个点进行测速。
结合上述第二方面第四种可能的实现方式，在第五种可能的实现方式中，该中央处理器用于对该第一数据进行测速处理和测角处理，以得到该第一雷达传感器的测速数据和测角数据，包括：该中央处理器用于获取第二频谱数据，该第二频谱数据为该第一频谱数据经过测距处理和测速处理后得到的频谱数据，其中，该第二频谱数据中包含至少一个第二频谱，该第二频谱为该第一频谱经过测距处理和测速处理后所得到的频谱；该中央处理器用于对该第二频谱数据中的所有第二频谱进行累加，其中，该累加可以是相干累加，也可以是非相干累加；在该第二频谱数据的所有该第二频谱中，该中央处理器用于对每个第二频谱同一位置的检测点进行快速傅里叶变换，从而得到每个检测点的测角数据。
本实施例中，第一微处理器对回波数据进行测距处理后得到第一频谱数据，中央处理器对第一频谱数据进行测速处理得到第二频谱数据，之后中央处理器对第二频谱数据中的所有第二频谱进行累加，对第二频谱同一位置的检测点进行快速傅里叶变换，从而得到每个检测点的测角数据；此种方式对所有检测点进行测距、测速及测角，实现对所有检测点的跟踪，之后再从跟踪的检测点中筛选出需要检测的目标，计算量较大，但本实施例中第 一微处理器与中央处理器协同工作的方式能够支撑该种工作方式所需要的算力。
结合上述第二方面第四种可能的实现方式，在第六种可能的实现方式中，该中央处理器用于对该第一数据进行测速处理和测角处理，以得到该第一雷达传感器的测速数据和测角数据，包括：该中央处理器用于获取第二频谱数据，该第二频谱数据为该第一频谱数据经过测距处理和测速处理后得到的频谱数据，其中，该第二频谱数据中包含至少一个第二频谱，该第二频谱为该第一频谱经过测距处理和测速处理后所得到的频谱；该中央处理器用于对该第二频谱数据中的所有该第二频谱进行累加，其中，该累加可以是相干累加，也可以是非相干累加；该中央处理器用于对累加后的该第二频谱数据进行二维恒虚警率，以从该检测点中获取第一目标检测点，该第一目标检测点为该检测点中的部分检测点；在该第二频谱数据的所有该第二频谱中，该中央处理器用于分别对每个第一目标检测点进行快速傅里叶变换，从而得到每个第一目标检测点的测角数据。
本实施例中，中央处理器对第一频谱数据进行测速处理后得到第二频谱数据，对第二频谱数据中的所有第二频谱进行累加，之后对累加后的第二频谱数据进行二维恒虚警率，以提取检测点中需要跟踪的部分检测点作为第一目标检测点，之后对每个第一目标检测点进行快速傅里叶变换，从而得到每个第一目标检测点的测角数据，此种方式先检测，之后通过二维恒虚警率筛选出需要跟踪的点进行跟踪，较为节省算力。
结合上述第二方面第二种可能的实现方式，在第七种可能的实现方式中，该第一射频单元还用于发送第一线性调频信号，该第一线性调频信号用于进行探测；该第一射频单元用于接收回波数据，包括：该第一射频单元用于接收第二线性调频信号，该第二线性调频信号为该第一射频单元发送该第一线性调频信号之后返回的线性调频信号。
本实施例中，第一射频单元发射的第一线性调频信号在触碰到障碍物之后反弹，从而使得第一射频单元能够接收到第二线性调频信号，从而能够进行探测。
结合上述第二方面第七种可能的实现方式，在第八种可能的实现方式中，该第一微处理器用于对该回波数据进行预处理，以得到第一数据，包括：该第一微处理器用于对该第二线性调频信号进行预处理，以得到测距数据；该中央处理器用于对该第一数据进行后处理，以得到该第一雷达传感器的第二数据，包括：该中央处理器用于，当该测距数据所对应的第二数字线性调频信号的数据量累积达到一帧时，对该第二数字线性调频信号进行测速处理和测角处理。
本实施例中，第一射频单元每接收到一个第二数字线性调频信号，第一微处理器对该第二数字线性调频信号执行一次测距处理，之后第一微处理器将雷达传感器的测距数据发送给中央处理器；由于需要多组测距数据才能进行测速，因此，当测距数据所对应的第二数字线性调频信号的数据量累积达到一帧时，中央处理器对第二数字线性调频信号进行测速处理和测角处理。
结合上述第二方面第七种可能的实现方式，在第九种可能的实现方式中，该第一微处理器用于对该回波数据进行预处理，以得到第一数据，包括：该第一微处理器用于对该第一射频单元所接收的X个该第二数字线性调频信号进行下采样，得到Y个第三数字线性调频信号，其中，X和Y均为大于1的正整数，其中，Y小于X；该中央处理器用于对该第一 数据进后处理，以得到该第一雷达传感器的第二数据，包括：该中央处理器用于对该第一数据中的第三数字线性调频信号进行测距处理、测速处理和测角处理，以得到测距数据、测速数据和测角数据。
本实施例中，当第一射频单元的采样量大于雷达系统所需的数量时，第一微处理器需要对第一射频单元所接收的第二数字线性调频信号进行下采样，从而压缩第一射频单元所接收的数据量，以满足雷达系统的工作要求。
结合上述第二方面第二至第九种可能的实现方式，在第十种可能的实现方式中，该第一微处理器用于对该回波数据进行预处理，以得到第一数据，包括；该第一微处理器用于对该回波数据进行压缩，以得到第一压缩数据；该中央处理器用于对该合并数据进行后处理，包括：该中央处理器用于对该压缩合并数据进行解压缩。
本实施例中，在第一微处理器对回波数据进行测距处理后，进一步对测距数据进行压缩，从而使得第一微处理器向中央处理器传输压缩后的数据，从而减少了第一微处理器和中央处理器之间的数据传输压力，提升了数据吞吐能力。
从以上技术方案可以看出，本申请实施例具有以下优点：
本发明实施例中，提供了一种雷达系统，可应用于自动驾驶的车载雷达，例如车载毫米波雷达，包括至少一个雷达传感器，该至少一个雷达传感器中的第一雷达传感器包括数据合并模块和M个雷达单片，其中，每个雷达单片包括第一射频单元和第一微处理器，该M为大于1的整数；该第一射频单元用于接收回波数据；该第一微处理器用于对该回波数据进行预处理，以得到第一数据；该数据合并模块用于对该M个雷达单片各自输出的第一数据进行合并，以得到该第一雷达传感器的合并数据；该数据合并模块还用于传输该合并数据，该合并数据用于经后处理后得到该第一雷达传感器的第二数据，该至少一个雷达传感器各自的第二数据用于生成该雷达系统的点云信息。本实施例中，第一微处理器与中央处理器通过任务分工，使系统能够兼容更大规模天线阵列的雷达传感器、能够适配不同的系统参数和算法处理流程；微处理器和中央处理器的算力负担、硬件资源负担都相应减小，资源分配更合理。同时，由于微处理器和中央处理器共同工作，提升了处理效率。
附图说明
图1为当前采用的分布式雷达信号处理单元的系统架构图；
图2为当前采用的单一中央雷达信号处理单元的系统架构；
图3为本发明实施例中雷达系统的系统架构图；
图4为本发明实施例中雷达系统的雷达传感器的一种实施方式的示意图；
图5为本发明实施例中雷达系统的雷达传感器的另一种实施方式的示意图；
图6为本发明实施例中雷达系统的雷达传感器的另一种实施方式的示意图；
图7为本发明实施例中雷达系统的一种实施方式的示意图；
图8为本发明实施例中雷达系统的另一种实施方式的示意图；
图9为一种雷达传感器的示意图；
图10为本发明实施例中雷达系统数据处理的流程图；
图11为本发明实施例中雷达系统处理数据时，对第二频谱进行测角处理的示意图。
具体实施方式
本发明实施例提供一种雷达系统，能够通过对雷达数据进行多级域的处理，从而调整雷达系统的分工，提升雷达系统的处理能力、数据缓存能力和数据吞吐能力。
为了使本技术领域的人员更好地理解本申请方案，下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚、完整地描述，显然，所描述的实施例仅仅是本申请一部分的实施例，而不是全部的实施例。基于本申请中的实施例，本领域普通技术人员在没有做出创造性劳动前提下所获得的所有其他实施例，都应当属于本申请保护的范围。
本申请的说明书和权利要求书及上述附图中的术语“第一”、“第二”、“第三”“第四”等(如果存在)是用于区别类似的对象，而不必用于描述特定的顺序或先后次序。应该理解这样使用的数据在适当情况下可以互换，以便这里描述的实施例能够以除了在这里图示或描述的内容以外的顺序实施。此外，术语“包括”和“具有”以及他们的任何变形，意图在于覆盖不排他的包含，例如，包含了一系列步骤或单元的过程、方法、系统、产品或设备不必限于清楚地列出的那些步骤或单元，而是可包括没有清楚地列出的或对于这些过程、方法、产品或设备固有的其它步骤或单元。
应当理解，在本申请中，“至少一个(项)”是指一个或者多个，“多个”是指两个或两个以上。“和/或”，用于描述关联对象的关联关系，表示可以存在三种关系，例如，“A和/或B”可以表示：只存在A，只存在B以及同时存在A和B三种情况，其中A，B可以是单数或者复数。字符“/”一般表示前后关联对象是一种“或”的关系。“以下至少一项(个)”或其类似表达，是指这些项中的任意组合，包括单项(个)或复数项(个)的任意组合。例如，a，b或c中的至少一项(个)，可以表示：a，b，c，“a和b”，“a和c”，“b和c”，或“a和b和c”，其中a，b，c可以是单个，也可以是多个。
高分辨率车载雷达是自动驾驶中不可缺少的重要组成部分，用于测量目标距离和速度等。在自动驾驶领域，雷达传感器需要提供更强的环境感知能力。增加雷达传感器阵列天线规模是一种典型的提高传感器对环境测量能力的方法。在测速和测距的基础上，更多的收发天线可以增强雷达传感器对目标角度的分辨能力。相应的，大规模天线阵列带来更强的数据吞吐需求、更多的数据存储需求以及更大的算力需求。因此，后续的信号处理和数据处理能力也需要匹配逐步提升的传感器规格。然而，车身可以安装雷达的位置以及相应的安装尺寸并没有持续增大，随着雷达传感器的天线阵列规模增大，接收天线数量大幅增加。
请参阅图1，图1为当前采用的分布式雷达信号处理单元的系统架构图，如图1所示，车体100周围配置6个雷达(也可以是多个，对比并不限定)，分别为前向雷达传感器101、后向雷达传感器102和四个角雷达传感器103,每个雷达传感器包含数字前端(digital front end,DFE)和传感器雷达信号处理单元(Sensor radar-signal processing unit)Sensor RPU两个部分。其中，DFE包含了收发天线及相应的射频前端电路，并通过模数转换器(analog-digital converter，ADC)将转换后的数字数据传输到Sensor RPU中进行处理。Sensor RPU进行毫米波雷达信号处理，通常包含测距、测速和测角等信号处理，也可能包含点云数据聚合以及动目标跟踪等高层次数据处理。经过每个雷达各自的数据处理， 将结果信息以数字信号的形式传输到中央雷达信号处理单元104(中央RPU)进行后续的多雷达数据融合等处理任务。
采用图1的架构搭建车载毫米波雷达系统，随着雷达传感器天线阵列规模逐渐增大、数据获取量和数据传输带宽逐渐增多，受限于体积、重量和功耗，各个雷达传感器会很快进入到处理能力、缓存能力和数据吞吐能力的瓶颈。因此难以适应更大天线阵列规模的雷达传感器处理需求，也难以兼容更灵活的系统参数设计。
请参阅图2，图2为当前采用的单一中央RPU的系统架构，如图2所示，车体200周围配置6个雷达(也可以是多个，对比并不限定)，分别为前向雷达传感器201、后向雷达传感器202和四个角雷达传感器203,每个雷达传感器。每个雷达仅包含模拟前端(analog front end,AFE)，AFE内部为收发天线和射频前端电路。接收天线获得的信号经过射频前端模拟电路放大、滤波、下混频等处理后，传输模拟信号到中央雷达信号处理单元204(中央RPU)。中央RPU204对应各个雷达的多个模拟信号通路在处理器内配备多个ADC。模拟信号经ADC转换后，在中央RPU204内部进行雷达数字信号处理以及高层数据融合处理。
采用图2的架构搭建车载毫米波雷达系统，随着雷达传感器的天线阵列规模增大，接收天线数量大幅增加，这将导致每个雷达传感器输出的模拟信号通道数大幅增加。相应的，在中央RPU204一端须配置数十个甚至上百个模拟信号接收端口以及对应的ADC，这将导致中央RPU204的芯片面积以及对应的处理板尺寸大幅增加。同时，由于信号处理与数据融合任务都由中央RPU204承担，对处理器算力带来巨大挑战。这些缺点也将制约系统对大规模天线阵列雷达传感器的适应性。
因此，对于多个雷达传感器的数据处理，若采用中央处理器统一处理的方式，将会给中央处理器的算力带来巨大挑战；若采用雷达传感器一端的处理器处理后发送给中央处理器进行汇总的方式，雷达传感器一端的处理器会很快进入到处理能力、缓存能力和数据吞吐能力的瓶颈；这些缺点将制约雷达系统对大规模天线阵列雷达传感器的适应性。
为解决上述问题，本申请实施例提供一种雷达系统，通过对雷达数据进行多级域的处理，从而调整雷达系统的分工，提升雷达系统的处理能力、数据缓存能力和数据吞吐能力。
需要说明的是，本申请所提供的雷达系统的使用场景不仅限于车载雷达，还可用于其他雷达系统，以及其他类似拓扑结构的电子处理系统，利用处理任务和处理模式的灵活配置使系统兼容多种外围配置。以标准化、小型化的分布式处理单元配合强大的中央处理单元实现系统的软升级。
为便于理解，以下结合附图，对本申请实施例所提供的雷达系统进行详细说明。
本申请所提供的雷达系统，包括至少一个雷达传感器，该至少一个雷达传感器中的第一雷达传感器包括数据合并模块和M个雷达单片，其中，每个雷达单片包括第一射频单元和第一微处理器，该M为大于1的整数；
该第一射频单元用于接收回波数据；
该第一微处理器用于对该回波数据进行预处理，以得到第一数据；
该数据合并模块用于对该M个雷达单片各自输出的第一数据进行合并，以得到该第一雷达传感器的合并数据；
该数据合并模块还用于传输该合并数据，该合并数据用于经后处理后得到该第一雷达传感器的第二数据，该至少一个雷达传感器各自的第二数据用于生成该雷达系统的点云信息。
本实施例中，数据合并模块传输的数据可以是发送给中央处理器，以下实施例中以下一级处理器为中央处理器为例进行说明，但是本申请实施例对此并不进行限定。当该雷达系统还包括中央处理器时，该数据合并模块还用于传输该合并数据，包括：该数据合并模块用于向该中央处理器传输该合并数据；该中央处理器用于对该合并数据进行后处理，以得到该第一雷达传感器的第二数据。
需要说明的是，上述数据合并模块对数据进行合并的具体实施方式，根据数据合并模块和下一级处理器(例如中央处理器)之间传输位宽的不同，可以分为数据打包和并行转串行两种方式，以下分别进行说明。
1、数据打包。
若数据合并模块与下一级处理器(例如中央处理器)之间的信号传输位宽大于或等于该第一雷达传感器的合并数据的位宽，则，该数据合并模块用于对该M个雷达单片各自输出的第一数据进行合并，包括：该数据合并模块用于对该M个雷达单片各自输出的第一数据进行打包。
本实施例中，由于数据合并模块与中央处理器之间的信号传输位宽大于或等于第一雷达传感器的合并数据的位宽，即，数据合并模块与中央处理器之间的传输信道能够完整地传输第一雷达传感器的合并数据，因此，数据合并模块只需要对M个雷达单片各自输出的第一数据进行打包，即可将所得到的合并数据发送给中央处理器。
2、并行转串行。
若数据合并模块与下一级处理器(例如中央处理器)之间的信号传输位宽小于该第一雷达传感器的合并数据的位宽，则，该数据合并模块用于对该M个雷达单片各自输出的第一数据进行合并，包括：该数据合并模块用于将该M个雷达单片各自输出的第一数据由并行数据转化为串行数据；该数据合并模块用于缓存该串行数据；该数据合并模块用于向该中央处理器传输该合并数据，包括：该数据合并模块用于分N次向该中央处理器传输该串行数据，该N为大于1的整数。
本实施例中，当数据合并模块与该中央处理器之间的信号传输位宽小于第一雷达传感器的合并数据的位宽时，数据合并模块无法通过传输通道一次性向中央处理器传输合并数据，因此，数据合并模块需要将合并数据由并行数据转化为串行数据并缓存，之后分次向中央处理器传输，从而使得传输位宽小于合并数据位宽时，数据合并模块能够向中央处理器传输完整的合并数据。
需要说明的是，该雷达系统的结构可以如图3所示，该雷达系统包括前向雷达传感器301、后向雷达传感器302和四个角雷达传感器303在，共六个雷达传感器，以及一个中央处理器304，其中，每个雷达传感器的结构相同，每个雷达传感器为级联的雷达传感器，分别包括M个雷达单片和数据合并模块，可选地，每个雷达传感器中，该数据合并模块可以为一个，也可以为多个，对此本申请实施例并不进行限定；可选地，雷达单片可以是通 过片上系统(system on chip,SoC)实现的，每个雷达单片分别包括射频单元和微处理器，其中，对于每个雷达单片中射频单元和微处理器的数量，本申请不进行限定。进一步地，该射频单元可以包括收发天线，例如多收多发的收发天线。如图4所示，作为一种举例，图4中的雷达传感器为四个SoC雷达单片401级联而成的雷达传感器，其中，每个SoC雷达单片401上设置有一个射频前端(RF前端)4011和一个微处理器4012，该RF前端4011提供RX1至RX4四个接收天线以及TX1和TX2两个发送天线，每个SoC雷达单片401构成2发4收的结构，四个SoC雷达单片401级联在一起构成8发16收的雷达天线阵列，四个SoC雷达单片401上的微处理器4011与同一个数据合并模块402连接，该数据合并模块402通过以太网向中央处理器发送合并数据。可选地，每个RF前端4011还可以包括功率放大器(power amplifier，PA)、低噪声放大器(low noise amplifier，LNA)、混频器、中频滤波器等RF电路以及线性调频信号chirp参数配置寄存器。
本实施例中，多片雷达级联构成的级联传感器具有较大的数据量，在现有技术的方案中，会对数据吞吐量和处理器的算力造成较大的挑战，为了解决此问题，本实施例所提供的方案在级联雷达传感器的每个雷达单片上设置微处理器，由微处理器对射频单元所获取的回波数据进行预处理，之后通过数据合并模块对雷达传感器中各个雷达单片预处理后的数据进行合并，得到合并数据，之后通过中央处理器对合并数据进行后处理，从而使得微处理器和中央处理器分担了算力，通过多级域的数据处理方式避免了级联雷达数据量增大给微处理器或中央处理器带来的算力负担。从而解决了大规模天线阵列雷达传感器中处理能力、缓存能力和数据吞吐能力的瓶颈问题，其中，该中央处理器可以是算力较强且接口丰富的多域控制器(multi domain controller,MDC)。
需要说明的是，上述中央处理器所进行的后处理，不是对各个雷达传感器处理好的数据进行融合，而是对第一雷达传感器预处理的数据进行进一步的处理，以得到该第一雷达传感器的第二数据，以下对预处理和后处理的详细流程进行说明。
1、第一微处理器用于对回波数据进行预处理，以得到第一数据，包括：
第一微处理器用于对所述回波数据进行测距处理，以得到测距数据。
可选地，该测距处理具体包括：
第一微处理器对回波数据进行数据补零；
第一微处理器对数据补零后的回波数据进行加窗处理；
第一微处理器对加窗后的回波数据进行快速傅里叶变换，以得到第一频谱数据，第一频谱数据中包含至少一个第一频谱。
本实施例中，由于第一射频单元可以为多发多收的结构，含有多个接收天线，因此，每个接收天线所接收的回波数据构成一个第一频谱。
进一步地，该第一频谱可以通过矩阵表示，该矩阵包括横向的行和纵向的列，其中，每一行记录了一个回波数据所检测的不同检测点，每一列记录了不同的回波数据所检测的同一检测点。其中，每个单频对应一个回波数据所检测的一个检测点，从而可以得到每个检测点的测距数据，该检测点为点云信息中的一个点。
进一步地，数据合并模块用于对M个雷达单片各自输出的第一数据进行合并，以得到 第一雷达传感器的合并数据，包括：
数据合并模块用于对M个雷达单片各自输出的测距数据进行合并，以得到第一雷达传感器的合并数据。
2、中央处理器对合并数据进行后处理，以得到第一雷达传感器的第二数据，包括：
中央处理器对合并数据进行测速处理和测角处理，以得到第一雷达传感器的测速数据和测角数据。
可选地，该测速处理的具体过程包括：
中央处理器根据所述第一频谱数据中每个单频所对应的距离数据，获取各个检测点的速度数据。
本实施例中，由于第一频谱数据中每个单频对应有一个检测点的测距数据，即距离数据，中央处理器在进行测速操作时，由于第一频谱的矩阵中每一列表示同一检测点在不同回波数据下的数据，因此对每一列做快速傅里叶变换，从而得到测速数据，此时得到第二频谱，该第二频谱中记录了每个检测点的测距数据和测速数据。
对于测角处理，可以分为先跟踪后检测(tracking before detection，TBD)，以及先检测后跟踪(tracking after detection，TAD)两种工作方式。以下分别进行说明。
1、先跟踪后检测(tracking before detection，TBD)。
中央处理器获取第二频谱数据，该第二频谱数据为第一频谱数据经过测距和测速处理后得到的频谱数据，其中该第二频谱数据中包含至少一个第二频谱，该第二频谱为第一频谱经过测距和测速处理后的所得到的频谱。
中央处理器对第二频谱数据中的所有第二频谱进行累加，该累加可以是相干累加或非相干累加。
在第二频谱数据的所有第二频谱中，中央处理器对同一位置的检测点进行快速傅里叶变换，从而得到每个检测点的测角数据。
本实施例中，中央处理器对每个点分别进行测角，从而能够得到每个点的距离、速度和角度数据；之后中央处理器筛选出距离、速度和角度稳定变化的点作为检测点并持续的检测。从而实现雷达系统的工作。由于本申请实施例中，中央处理器所接收的合并数据是经过了微处理进行测距处理的，因此中央处理器的算力能够先检测，获得每个检测点距离和速度变化的趋势，从而从所有检测点中获取速度和距离线性变化的检测点作为第一目标检测点，再对第一目标检测点进行跟踪，因此能够确保获得最充分的点云信息，以提升雷达检测的精确度。
2、先检测后跟踪(tracking after detection，TAD)。
中央处理器获取第二频谱数据；
中央处理器对第二频谱数据中的所有第二频谱进行累加，该累加可以是相干累加或非相干累加。
中央处理器对累加后的第二频谱数据进行二维恒虚警率，以从检测点中获取第一目标检测点，该第一目标检测点为该检测点中的部分检测点；
在第二频谱数据的所有第二频谱中，中央处理器分别对每个第一目标检测点进行快速 傅里叶变换，从而得到每个第一目标检测点的测角数据。
本实施例中，中央处理器对合并数据中的第一频谱数据进行相干累加或非相干累加，以进行二维恒虚警率(constant false alarm ratio,CFAR),通过对2维-CFAR的解读获取第一目标检测点，之后通过分别对第二频谱数据中的每个第一目标检测点进行角度快速傅里叶变换，得到第一目标检测点的角度数据。本实施例通过2维-CFAR的方式，先检测出第一目标检测点，再对该第一目标检测点进行跟踪，从而能够节省中央处理器的算力。
实际工作时，本申请实施例能够支持TAD或TBD两种工作方式，本领域技术人员可以根据需要进行选择，对此本申请实施例并不进行限定。
需要说明的是，上述的回波数据，可以是通过线性调频信号chirp信号来实现的，以下对chirp信号在本申请实施例所提供的雷达系统中的具体工作方式做详细说明。
在上述第一射频单元接收回波数据之前，还包括：
第一射频单元发送第一线性调频信号。
本实施例中，第一线性调频信号为用于进行探测的信号。
则，该第一射频单元接收回波数据具体为：
第一射频单元用于第二线性调频信号。
本实施例中，第二线性调频信号为所述第一射频单元发送第一线性调频信号之后，第一线性调频信号探测到被探测物后回波反弹的信号。
则，上述步骤中所述的对回波数据的处理，具体为对该第二线性调频信号的处理。其中，M个雷达单片中的每个雷达单片每输出一个第二数字线性调频信号的测距数据，对M个雷达单片各自输出的测距数据进行合并，以得到第一雷达传感器的合并数据，当测距数据所对应的第二数字线性调频信号的数据量累积达到一帧时，对第二数字线性调频信号进行测速处理和测角处理。
需要说明的是，若干个第二数字线性调频信号构成一帧第二数字线性调频信号，具体一帧包含多少个第二数字线性调频信号，取决于雷达系统的分辨率，雷达系统的分辨率越高，一帧所包含的第二数字线性调频信号数量越多，本领域技术人员可根据实际工作需求，设定一帧第二数字线性调频信号所对应的第二数字线性调频信号个数。例如，chirp信号持续时间为15us，为了测量速度，每帧积累64个chirp信号，那么原始回波数据大小约为8M字节。由于中央处理器无法仅凭一组测距数据确定检测点的速度，因此，中央处理器需要等到所接收的第二线性调频信号达到一帧时，才开始进行测速处理和测角处理。
以下通过举例对雷达系统对线性调频信号的具体处理方式做详细说明。
雷达系统的射频单元通过发送天线发送第一线性调频信号，之后通过接收天线接收第一线性调频信号回波之后的第二线性调频信号，假设一个发送天线在一帧内发送64个第一线性调频信号chirps信号，每个chirps信号检测1024个检测点，则接收天线同样会接收到一帧内的64个chirps信号，每个chirps信号检测1024个检测点，对接收天线所接收到的chirps信号进行归纳后，得到第一频谱，该第一频谱可以表示为如下公式1所示的第一二维矩阵：
Figure PCTCN2019092084-appb-000001
如公式1所示，第二线性调频信号中，每个chirps信号的每个检测点构成一个样本(Sample，S)，在公式1中，S 1,1表示第一个chirps信号的第一个检测点，S 1,1024表示第一个chirps信号的第1024个检测点，依次类推，S 64,1024表示第64个chirps信号的第1024个检测点。
基于上述公式1，本申请实施例所提供的雷达系统的具体处理流程如下：
1、测距处理。
本实施例中，第一微处理器对回波数据进行预处理，以得到第一数据的过程中。包括：
在第一微处理器获取回波数据的过程中，每获取一个第二线性调频信号即可获得上述公式1中矩阵的一个行，例如第二线性调频信号中的第一个chirps信号S1,第一微处理器实时对S1做快速傅里叶变换，以得到测距数据S 1,1至S 1,1024，从而完成第二线性调频信号中的第一个chirps信号的测距处理，得到上述公式1中的第一行测距数据，之后通过数据合并模块的合并，将该第一行测距数据发送给中央处理器。
2、测速处理。
中央处理器对合并数据进行测速处理，以得到第一雷达传感器的测速数据，包括：
当中央处理器所接收到的第二线性调频信号累积到一帧时，中央处理器得到上述公式1所示的第一二维矩阵的全部行的数据。中央处理器对第一二维矩阵的每一列的检测点做快速傅里叶变化，从而得到每个检测点的速度信息，从而完成测速处理，例如对S 1,1至S 64,1这一行做快速傅里叶变化，得到RD 1,1至RD 64,1，从而得到第二频谱，该第二频谱可以表示为以下第二二维矩阵：
Figure PCTCN2019092084-appb-000002
如公式2所示，在第二二维矩阵中，每个检测点S经过两次快速傅里叶变换，得到检测点RD，其中，每个检测点RD即为一个单频，每个单频记录有该检测点的测距数据和测速数据，例如，RD 1，1表示该检测点的距离为3m，速度为2m/s；RD 1，2表示该检测点的距离为4m，速度为5m/s；RD 1，1024表示该检测点的距离为1024m，速度为5m/s。
3、测角处理。
请参阅图11，本申请实施例所提供的雷达系统中，射频单元可以为多发多收的结构，含有多个接收天线，因此，每个接收天线所接收的回波数据构成一个第一频谱，相应地经过上述测速处理后得到多个第二频谱，如图11所示，以2发4收的射频单元为例，由于具有4个接收天线，则每个第二线性调频信号会被四个接收天线接收，相应地生成RD-map1至RD-map4四个第二频谱。
如上所述，对于第二频谱数据的测角处理可以分为TAD或TBD两种工作方式，以下具体进行说明。
(1)、TBD。
中央处理器对全部四个第二频谱进行累加，得到如图11所示的第二频谱数据。
在第二频谱数据的所有第二频谱中，中央处理器对同一位置的检测点进行快速傅里叶变换，从而得到每个检测点的测角数据。
本实施例中，如图11所示，以RD 1,1为例，每个第二频谱RD-map上的检测点RD 1,1记录了第二线性调频信号中第一个chirps信号在第一个检测点上的距离信息和速度信息；由于四个接收天线生成了四个RD-map，因此每个RD-map中分别包含一个RD 1,1，对四个RD-map分别对应的四个RD 1,1做快速傅里叶变化，从而得到RD 1,1的角度信息，从而最终获取第一个检测点的距离、速度和角度。同理，对RD-map上的每个检测点进行同样的处理，即可得到全部检测点的距离、速度和角度。最后中央处理器从中筛选出距离、速度和角度稳定变化的检测点进行持续，从而实现雷达系统的工作。
(2)、TAD。
中央处理器对第二频谱数据中的所有第二频谱进行二维恒虚警率(constant false alarm ratio,CFAR)，以获取第一目标检测点，第一目标检测点为距离和速度稳定变化的检测点。
本实施例中，由于第二频谱数据中本身就记录有每个检测点的距离信息和速度信息，因此，只需要对第二频谱数据中的所有第二频谱进行2维-CFAR处理，即可获得距离和速度稳定变化的检测点。由于在雷达系统中，只有距离和速度稳定变化的检测点才具有检测的意义，因此可以通过此种方式预先筛掉不需要检测的点。
中央处理器对第二频谱数据中的所有第二频谱进行累加。
本实施例中，该累加可以是相干累加或非相干累加。累加后得到如图11所示的第二频谱数据。
在第二频谱数据的所有第二频谱中，中央处理器分别对每个第一目标检测点进行快速傅里叶变换，从而得到每个第一目标检测点的测角数据。
本实施例中，仍然以如图11所示，需要说明的是，虽然TBD与TAD两种工作方式中，供中央处理器进行测角处理的第二频谱数据都可以用图11来表示，但是二者具有区别：在TBD的工作方式中，RD-map1至RD-map4中每个RD-map均包含全部检测点的信息；而在TAD的工作方式中，RD-map1至RD-map4中每个RD-map中包含第一目标检测点的信息。
进一步地，在TAD的工作方式中，假设以RD 1,1为其中一个第一目标检测点，每个第二 频谱RD-map上的检测点RD 1,1记录了第二线性调频信号中第一个chirps信号在第一个检测点上的距离信息和速度信息；由于四个接收天线生成了四个RD-map，因此每个RD-map中分别包含一个RD 1,1，对四个RD-map分别对应的四个RD 1,1做快速傅里叶变化，从而得到RD 1,1的角度信息，从而最终获取第一个第一目标检测点的距离、速度和角度。同理，对RD-map上的每个第一目标检测点进行同样的处理，即可得到全部第一目标检测点的距离、速度和角度。从而实现雷达系统的工作。
需要说明的是，为了进一步提升微处理器和中央处理器之间的数据吞吐能力，微处理器对预处理后的数据进行压缩，中央处理器对收到的合并数据进行解压，从而能够减少数据传输的数据量，提升数据吞吐能力。为便于理解，以下对压缩解压的具体流程进行详细说明。
上述第一微处理器对回波数据进行预处理，以得到第一数据，包括；
第一微处理器对回波数据进行压缩，以得到第一压缩数据。
本实施例中，第一微处理器在完成了对回波数据的处理后，才进行压缩，其中，第一微处理器对回波数据的处理可以是上述的任意一种处理方式，可参见前文的描述进行理解，此处不再赘述。
上述数据合并模块用于对M个雷达单片各自输出的第一数据进行合并，以得到第一雷达传感器的合并数据，包括：
数据合并模块用于对M个雷达单片各自输出的第一压缩数据进行合并，以得到第一雷达传感器的压缩合并数据。
本实施例中，数据合并模块对压缩后的数据进行合并，以得到压缩合并数据。
中央处理器用于对合并数据进行后处理，包括：
中央处理器用于对压缩合并数据进行解压缩。
本实施例中，中央处理器在收到压缩合并数据后，对其进行解压，以进行后续步骤。
需要说明的是，上述压缩解压步骤所采用的技术方案可以为任意压缩解压方案，例如，数据压缩算法可以依赖于距离快速傅里叶变化结果信息或者原始回波数据时域信息，实现有损压缩或无损压缩，对此本申请实施例并不进行限定。
进一步地，第一微处理器在开始处理回波数据之前，还可以对回波数据进行下采样，从而进一步压缩需要处理的数据量，具体下采样的采样数据量可以根据实际使用需求而定，以下以回波数据为第二数字线性调频信号为例，对下采样的工作方式进行详细说明。
上述第一微处理器对回波数据进行预处理，以得到第一数据，包括：
第一微处理器对第一射频单元所接收的X个第二数字线性调频信号进行下采样，得到Y个第三数字线性调频信号，其中，X和Y均为大于1的正整数，其中，Y小于X。
本实施例中，对于第一射频单元所接收的第二数字线性调频信号，第一微处理器只针对其中的一部分进行抽样处理，从而减少了需要处理的数据量。
数据合并模块用于对M个雷达单片各自输出的第一数据进行合并，以得到第一雷达传感器的合并数据，包括：
数据合并模块用于对M个雷达单片各自输出的第三数字线性调频信号进行合并，以得 到第一雷达传感器的合并数据。
本实施例中，M个雷达单片各自输出的第三数字线性调频信号为经过预处理后的第三数字线性调频信号。
中央处理器用于对合并数据进后处理，以得到第一雷达传感器的第二数据，包括：
中央处理器用于对合并数据中的第三数字线性调频信号进行后处理，以得到第一雷达传感器的第二数据。
需要说明的是，当雷达系统的工作流程包含下采样的技术方案时，可以包括以下4种处理方式。以下详细进行说明。
1、微处理下采样-微处理器测距-数据合并模块合并数据-中央处理器测速、测角。
本实施例中，微处理器对射频单元所接收的线性调频信号进行下采样，之后微处理器对下采样后的数据进行测距操作，经数据合并单元合并发送后，中央处理器对合并数据进行测速和测角操作。所述的下采样、测距、测速及测角操作的具体实现方式可参考上述记载，此处不再赘述。
2、微处理下采样-微处理器测距-微处理器压缩数据-数据合并模块合并数据-中央处理器解压、测速、测角。
本实施例中，微处理器对射频单元所接收的线性调频信号进行下采样，之后微处理器对下采样后的数据进行测距处理，之后微处理器对测距处理后的数据进行压缩，压缩数据经数据合并单元合并发送后，中央处理器对合并数据进行解压、测速和测角操作。所述的下采样、测距、测速及测角操作的具体实现方式可参考上述记载，此处不再赘述。
3、微处理下采样-数据合并模块合并数据-中央处理器测距、测速、测角。
本实施例中，微处理器仅仅对线性调频信号进行下采样，之后将下采样之后的数据通过数据合并模块发送给中央处理器进行测距、测速及测角处理，由于线性调频信号经历了下采样的步骤，数据量减小，因此同样能够解决数据吞吐瓶颈的问题，上述下采样、测距、测速及测角处理的具体实现方式可参考上述记载，此处不再赘述。
4、微处理下采样-微处理压缩数据-数据合并模块合并数据-中央处理器解压、测距、测速、测角。
本实施例中，微处理器仅仅对线性调频信号进行下采样，并对下采样之后的数据进行压缩，之后将压缩后的数据通过数据合并模块发送给中央处理器进行解压测距、测速及测角处理，由于线性调频信号经历了下采样的步骤，数据量减小，同时，对下采样之后的数据进行压缩，进一步解决了数据吞吐瓶颈的问题，上述下采样、测距、测速及测角处理的具体实现方式可参考上述记载，此处不再赘述。
本申请实施例还提供另一种雷达系统，本雷达系统与上述雷达系统的区别在于雷达传感器，每个雷达传感器包括微处理器和至少一个射频单元，该雷达传感器具体可以有以下2种实现方式：
1、多片小规模雷达前端级联的雷达传感器。
请参阅图5，如图5所示，该雷达传感器包括一个微处理器501和4个雷达单片502，其中，该4个雷达单片502全部与同一微处理器501连接，由该微处理器501来对4个雷 达单片502所获取到的回波数据来进行处理。可选地，该雷达传感器也可以是至少一个雷达单片502和一个微处理器组，该微处理器组内包括至少一个微处理器501，该微处理器组用于对至少一个雷达单片502的回波数据进行处理。进一步地，该雷达单片502包括射频单元、功率放大器(power amplifier，PA)、低噪声放大器(low noise amplifier，LNA)、混频器、中频滤波器IF等RF电路以及线性调频信号chirp参数配置寄存器，其中，该射频单元可以包括收发天线，例如可以包括RX1至RX4四个接收天线以及TX1和TX2两个发送天线，每个雷达单片502构成2发4收的结构，四个雷达单片502级联在一起构成8发16收的雷达天线阵列，其中，RX1至RX4分别连接一个LNA，每个LNA通过各自的混频器依次连接IF和ADC，最终每个雷达单片502的ADC与同一微处理器501连接。
2、接收端与发送端分置的雷达传感器。
请参阅图6，如图6所示，该雷达传感器包括一个微处理器以及与该微处理器连接的射频单元，其中，该射频单元包括分离设置的一个接收端射频前端(RX RF前端)和一个发送端射频前端(TX RF前端)，其中，该RX RF前端通过ADC与微处理器连接并向微处理器发送回波数据，TX RF前端通过压控振荡器和Chirp参数配置器与微处理器连接，并接受微处理器发送的信号。
基于上述雷达传感器的不同，本申请提供的另一种雷达系统与前述雷达系统的区别在于：由于射频单元采用多对一的方式或一对一的方式与微处理器连接，因此，不再需要数据合并单元来对各个射频单元所获取的回波数据进行合并，而是直接由微处理器来接收各个射频单元发送的回波数据，并由微处理器来将预处理后的回波数据发送给中央处理器。
进一步地，本申请所提供的另一种雷达系统的其余步骤可以参阅上述记载的内容，对此本申请实施例不再赘述。
上述对本申请实施例所提供的两种雷达系统的工作原理做了详细说明，在实际使用过程中，根据使用场景和需求的不同，本申请实施例所提供的雷达系统在车载雷达领域可以有以下几种具体的实现方式，以下结合附图进行详细说明。
1、车体周围布置同样的雷达。
请参阅图7，如图7所示，本申请第一种具体的实施方式所提供的车载雷达系统，包括车体700、中央处理器704以及分别设置在车体正面的前向雷达传感器701、背面的后向雷达传感器702以及四个角的角雷达传感器703，共6和雷达传感器；该前向雷达传感器701、后向雷达传感器702以及角雷达传感器703分别与该中央处理器704连接，其中，上述6个雷达传感器的结构相同，可以为上述任意雷达系统所述的雷达传感器，关于雷达传感器的具体结构，以及雷达传感器与中央处理器704配合工作处理回波数据的具体工作方式，可参阅上述记载的内容，此处不再赘述。
进一步地，该中央处理器704可以为中央雷达信号处理单元(central radar-signal processing unit,central RPU)，对于该雷达传感器的具体数量以及设置位置，本领域技术人员可以在实际使用过程中根据需要进行调整，对此本申请实施例并不进行限定。
本实施例中，通过在车体周围设置本申请实施例所提供的雷达传感器，并将该雷达传感器与中央处理器连接，以实现本申请实施例所提供的多级域数据方法，从而使得雷达传 感器中的微处理器和中央处理器分担了数据处理的算力压力，使得雷达系统的处理能力、缓存能力和数据吞吐能力得到提升，以适应大规模天线阵列雷达传感器所带来的巨大数据量。
需要说明的是，对于一些应用场景，对雷达于系统中的前向和后向雷达传感器需要较大的探测范围以及较高的点云密度，而对角雷达仅需要中等的探测范围以及点云密度。因此，可以对雷达系统中不同探测方位的雷达传感器做差异化的处理，以下针对此种情况进行具体的说明。
2、车体周围布置不同的雷达。
请参阅图8，如图8所示，本申请第二种具体的实施方式所提供的车载雷达系统，包括车体800、中央处理器804以及分别设置在车体正面的前向雷达传感器801、背面的后向雷达传感器802以及四个角的角雷达传感器803，共6和雷达传感器；该前向雷达传感器801、后向雷达传感器802以及角雷达传感器803分别与该中央处理器804连接，其中，该前向雷达传感器801及该后向雷达传感器802为本申请实施例所提供的雷达传感器，具体结构可参阅上述记载，此处不再赘述；该四个角雷达传感器803为现有技术中的雷达，如图9所示，具体可以包括一个雷达单片，该雷达单片中包括射频单元、功率放大器(power amplifier，PA)、低噪声放大器(low noise amplifier，LNA)、混频器、中频滤波器IF等RF电路以及线性调频信号chirp参数配置寄存器，其中，该射频单元可以包括收发天线，例如可以包括RX1至RX4四个接收天线以及TX1和TX2两个发送天线，2发4收的结构，其中，RX1至RX4中的每个接收天线分别连接一个LNA，每个LNA通过各自的混频器依次连接IF和ADC，最终与同一微处理器连接。由于角雷达传感器仅需要中等的探测范围以及点云密度，因此数据量较小，该微处理器可以独立处理该雷达单片的回波数据，包括测距处理、测速处理及测角处理，之后将处理好的数据发送给中央处理器804。
基于上述架构，图8所提供的车载雷达系统中，前向雷达传感器和后向雷达传感器使用本申请实施例所提供的雷达传感器，角雷达使用现有技术中的雷达传感器，请参阅图10，图8所提供的雷达系统中，中央处理器对于数据的处理包含以下工作步骤。
1001、中央处理器接收雷达传感器发来的数据。
本实施例中，该数据可以是雷达传感器的数据合并模块发来的合并数据，也可以是雷达传感器的微处理器直接发来的数据，具体工作方式可参阅上述记载，此处不再赘述。
进一步地，该中央处理器可以是通过以太网接收雷达传感器发来的数据，
1002、中央处理器判断所接收的数据是否来自前向雷达传感器或后向雷达传感器。
本实施例中，雷达传感器在向中央处理器发送数据时，会给发送的数据打上标签，从而中央处理器可以根据该标签，判断发送该数据的雷达传感器。
若中央处理器判断所接收的数据不是前向雷达传感器或后向雷达传感器，则执行步骤1003至1004。
1003、中央处理器获取该数据中的点云数据。
本实施例中，中央处理器所接收的数据为角雷达传感器处理好的数据，因此中央处理器可以直接获取到点云数据。
1004、中央处理器根据点云数据实现动目标跟踪。
本实施例中，中央处理器根据点云数据实现动目标跟踪从而完成对所扫描到的目标的跟踪。
若中央处理器判断所接收的数据是前向雷达传感器或后向雷达传感器，则执行步骤1005。
1005、中央处理器对数据进行处理。
本实施例中，中央处理器对数据的处理可以为本申请实施例所提供的雷达系统中，所公开的任意一种中央处理器的处理方式，可参数上述记载进行理解，此处不再赘述。
本实施例中，对于前向和后向雷达传感器，中央处理器采用本申请实施例所提供的处理流程进行后续处理。对于角雷达传感器，中央处理器则可以获取点云数据并进行后续的动目标跟踪处理。在对于不同的雷达传感器数据进行差异化处理后，中央处理器进行360度雷达数据融合以及异构传感器数据融合。从而能够根据实际需求，对设置在车体周围不同的雷达传感器进行处理，能够适应不同的使用需求，同时节省了算力。
需要说明的是，本申请实施例所提供的技术方案除了雷达系统外，也可用于其他类似拓扑结构的电子处理系统，利用处理任务和处理模式的灵活配置使系统兼容多种外围配置。以标准化、小型化的分布式处理单元配合强大的中央处理单元实现系统的软升级。
所属领域的技术人员可以清楚地了解到，为描述的方便和简洁，上述描述的系统的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。
在本申请所提供的几个实施例中，应该理解到，所揭露的系统可以通过其它的方式实现。例如，以上所描述的装置实施例仅仅是示意性的，例如，所述单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通信连接可以是通过一些接口，装置或单元的间接耦合或通信连接，可以是电性，机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外，在本申请各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现，也可以采用软件功能单元的形式实现。
所述集成的单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读取存储介质中。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分或者该技术方案的全部或部分可以以软件产品的形式体现出来，该计算机软件产品存储在一个存储介质中，包括若干指令用以使得一台计算机设备(可以是个人计算机，服务器，或者网络设备等)执行本申请各个实施例所述方法的全部或部分步骤。而前述的存储介质包括：U盘、移动硬盘、只读存储器(ROM，Read-Only Memory)、随机存取存储器(RAM，Random Access Memory)、磁碟或者光盘等各种可以存储程序代码 的介质。
以上对本发明实施例所提供的雷达系统进行了详细介绍，本文中应用了具体个例对本发明的原理及实施方式进行了阐述，以上实施例的说明只是用于帮助理解本发明的方法及其核心思想；同时，对于本领域的一般技术人员，依据本发明的思想，在具体实施方式及应用范围上均会有改变之处，综上所述，本说明书内容不应理解为对本发明的限制。

Claims (24)

  1. 一种雷达系统，其特征在于，包括至少一个雷达传感器，所述至少一个雷达传感器中的第一雷达传感器包括数据合并模块和M个雷达单片，其中，每个雷达单片包括第一射频单元和第一微处理器，所述M为大于1的整数；
    所述第一射频单元用于接收回波数据；
    所述第一微处理器用于对所述回波数据进行预处理，以得到第一数据；
    所述数据合并模块用于对所述M个雷达单片各自输出的第一数据进行合并，以得到所述第一雷达传感器的合并数据；
    所述数据合并模块还用于传输所述合并数据，所述合并数据用于经后处理后得到所述第一雷达传感器的第二数据，所述至少一个雷达传感器各自的第二数据用于生成所述雷达系统的点云信息。
  2. 根据权利要求1所述的雷达系统，其特征在于，还包括中央处理器，所述数据合并模块还用于传输所述合并数据，包括：
    所述数据合并模块用于向所述中央处理器传输所述合并数据；
    所述中央处理器用于对所述合并数据进行后处理，以得到所述第一雷达传感器的第二数据。
  3. 根据权利要求2所述的雷达系统，其特征在于，所述数据合并模块与所述中央处理器之间的信号传输位宽大于或等于所述第一雷达传感器的合并数据的位宽，则，所述数据合并模块用于对所述M个雷达单片各自输出的第一数据进行合并，包括：
    所述数据合并模块用于对所述M个雷达单片各自输出的第一数据进行打包。
  4. 根据权利要求2所述的雷达系统，其特征在于，所述数据合并模块与所述中央处理器之间的信号传输位宽小于所述第一雷达传感器的合并数据的位宽，则，所述数据合并模块用于对所述M个雷达单片各自输出的第一数据进行合并，包括：
    所述数据合并模块用于将所述M个雷达单片各自输出的第一数据由并行数据转化为串行数据；
    所述数据合并模块用于缓存所述串行数据；
    所述数据合并模块用于向所述中央处理器传输所述合并数据，包括：
    所述数据合并模块用于分N次向所述中央处理器传输所述串行数据，所述N为大于1的整数。
  5. 根据权利要求2至4任一所述的雷达系统，其特征在于，所述第一微处理器用于对所述回波数据进行预处理，以得到第一数据，包括：
    所述第一微处理器用于对所述回波数据进行测距处理，以得到测距数据；
    所述数据合并模块用于对所述M个雷达单片各自输出的第一数据进行合并，以得到所述第一雷达传感器的合并数据，包括：
    所述数据合并模块用于对所述M个雷达单片各自输出的测距数据进行合并，以得到所述第一雷达传感器的合并数据；
    所述中央处理器用于对所述合并数据进行后处理，以得到所述第一雷达传感器的第二 数据，包括：
    所述中央处理器用于对所述合并数据进行测速处理和测角处理，以得到所述第一雷达传感器的测速数据和测角数据。
  6. 根据权利要求5所述的雷达系统，其特征在于，所述第一微处理器用于对所述回波数据进行测距处理，以得到测距数据，包括：
    所述第一微处理器用于对所述回波数据进行数据补零；
    所述第一微处理器还用于对数据补零后的所述回波数据进行加窗处理；
    所述第一微处理器还用于对加窗后的所述回波数据进行快速傅里叶变换，以得到第一频谱数据，所述第一频谱数据中包含至少一个第一频谱，其中，所述第一频谱中的每个单频对应一个检测点的距离数据，所述检测点为所述点云信息中的一个点。
  7. 根据权利要求6所述的雷达系统，其特征在于，所述中央处理器用于对所述合并数据进行测速处理和测角处理，以得到所述第一雷达传感器的测速数据和测角数据，包括：
    所述中央处理器用于根据所述第一频谱数据中每个单频所对应的距离数据，获取各个检测点的测速数据。
  8. 根据权利要求7所述的雷达系统，其特征在于，所述中央处理器用于对所述合并数据进行测速处理和测角处理，以得到所述第一雷达传感器的测速数据和测角数据，包括：
    所述中央处理器用于获取第二频谱数据，所述第二频谱数据为所述第一频谱数据经过测距处理和测速处理后得到的频谱数据，其中，所述第二频谱数据中包含至少一个第二频谱，所述第二频谱为所述第一频谱经过测距处理和测速处理后所得到的频谱；
    所述中央处理器用于对所述第二频谱数据中的所有第二频谱进行累加；
    在所述第二频谱数据的所有所述第二频谱中，所述中央处理器用于对每个第二频谱同一位置的检测点进行快速傅里叶变换，从而得到每个检测点的测角数据。
  9. 根据权利要求7所述的雷达系统，其特征在于，所述中央处理器用于对所述合并数据进行测速处理和测角处理，以得到所述第一雷达传感器的测速数据和测角数据，包括：
    所述中央处理器用于获取第二频谱数据，所述第二频谱数据为所述第一频谱数据经过测距处理和测速处理后得到的频谱数据，其中，所述第二频谱数据中包含至少一个第二频谱，所述第二频谱为所述第一频谱经过测距处理和测速处理后所得到的频谱；
    所述中央处理器用于对所述第二频谱数据中的所有所述第二频谱进行累加；
    所述中央处理器用于对累加后的所述第二频谱数据进行二维恒虚警率，以从所述检测点中获取第一目标检测点，所述第一目标检测点为所述检测点中的部分检测点；
    在所述第二频谱数据的所有所述第二频谱中，所述中央处理器用于分别对每个第一目标检测点进行快速傅里叶变换，从而得到每个第一目标检测点的测角数据。
  10. 根据权利要求2至4任一所述的雷达系统，其特征在于，所述第一射频单元还用于发送第一线性调频信号，所述第一线性调频信号用于进行探测；
    所述第一射频单元用于接收回波数据，包括：
    所述第一射频单元用于接收第二线性调频信号，所述第二线性调频信号为所述第一射频单元发送所述第一线性调频信号之后返回的线性调频信号。
  11. 根据权利要求10所述的雷达系统，其特征在于，
    所述第一微处理器用于对所述回波数据进行预处理，以得到第一数据，包括：
    所述第一微处理器用于对所述第二线性调频信号进行预处理，以得到测距数据；
    所述数据合并模块用于对所述M个雷达单片各自输出的测距数据进行合并，以得到所述第一雷达传感器的合并数据，包括：
    所述数据合并模块用于，所述M个雷达单片中的每个雷达单片每输出一个第二数字线性调频信号的测距数据，对M个雷达单片各自输出的测距数据进行合并，以得到所述第一雷达传感器的合并数据；
    所述中央处理器用于对所述合并数据进行后处理，以得到所述第一雷达传感器的第二数据，包括：
    所述中央处理器用于，当所述测距数据所对应的第二数字线性调频信号的数据量累积达到一帧时，对所述第二数字线性调频信号进行测速处理和测角处理。
  12. 根据权利要求10所述的雷达系统，其特征在于，所述第一微处理器用于对所述回波数据进行预处理，以得到第一数据，包括：
    所述第一微处理器用于对所述第一射频单元所接收的X个所述第二数字线性调频信号进行下采样，得到Y个第三数字线性调频信号，其中，X和Y均为大于1的正整数，其中，Y小于X；
    所述数据合并模块用于对所述M个雷达单片各自输出的第一数据进行合并，以得到所述第一雷达传感器的合并数据，包括：
    所述数据合并模块用于对所述M个雷达单片各自输出的第三数字线性调频信号进行合并，以得到所述第一雷达传感器的合并数据；
    所述中央处理器用于对所述合并数据进后处理，以得到所述第一雷达传感器的第二数据，包括：
    所述中央处理器用于对所述合并数据中的第三数字线性调频信号进行后处理，以得到所述第一雷达传感器的第二数据。
  13. 根据权利要求1至12任一所述的雷达系统，其特征在于，所述第一微处理器用于对所述回波数据进行预处理，以得到第一数据，包括；
    所述第一微处理器用于对所述回波数据进行压缩，以得到第一压缩数据；
    所述数据合并模块用于对所述M个雷达单片各自输出的第一数据进行合并，以得到所述第一雷达传感器的合并数据，包括：
    所述数据合并模块用于对所述M个雷达单片各自输出的所述第一压缩数据进行合并，以得到所述第一雷达传感器的压缩合并数据；
    所述中央处理器用于对所述合并数据进行后处理，包括：
    所述中央处理器用于对所述压缩合并数据进行解压缩。
  14. 一种雷达系统，其特征在于，包括至少一个雷达传感器，所述至少一个雷达传感器中的第一雷达传感器包括第一微处理器和至少一个第一射频单元；
    所述至少一个第一射频单元用于接收至少一个回波数据；
    所述第一微处理器用于对所述至少一个回波数据进行预处理，以得到第一数据；
    所述第一微处理器还用于传输所述第一数据，所述第一数据用于经后处理后得到所述第一雷达传感器的第二数据，所述至少一个雷达传感器各自的第二数据用于生成所述雷达系统的点云信息。
  15. 根据权利要求14所述的雷达系统，其特征在于，还包括中央处理器，所述第一微处理器还用于传输所述第一数据，包括：
    所述第一微处理器用于向所述中央处理器传输所述第一数据；
    所述中央处理器用于对所述第一数据进行后处理，以得到所述第一雷达传感器的第二数据。
  16. 根据权利要求15所述的雷达系统，其特征在于，所述第一微处理器用于对所述回波数据进行预处理，以得到第一数据，包括：
    所述第一微处理器用于对所述回波数据进行测距处理，以得到测距数据；
    所述中央处理器用于对所述第一数据进行后处理，以得到所述第一雷达传感器的第二数据，包括：
    所述中央处理器用于对所述第一数据进行测速处理和测角处理，以得到所述第一雷达传感器的测速数据和测角数据。
  17. 根据权利要求16所述的雷达系统，其特征在于，所述第一微处理器用于对所述回波数据进行测距处理，以得到测距数据，包括：
    所述第一微处理器用于对所述回波数据进行数据补零；
    所述第一微处理器还用于对数据补零后的所述回波数据进行加窗处理；
    所述第一微处理器还用于对加窗后的所述回波数据进行快速傅里叶变换，以得到第一频谱数据，所述第一频谱数据中包含至少一个第一频谱，其中，所述第一频谱中的每个单频对应一个检测点的距离数据，所述检测点为所述点云信息中的一个点。
  18. 根据权利要求17所述的雷达系统，其特征在于，所述中央处理器用于对所述第一数据进行测速处理和测角处理，以得到所述第一雷达传感器的测速数据和测角数据，包括：
    所述中央处理器用于根据所述第一频谱数据中每个单频所对应的距离数据，获取各个检测点的测速数据。
  19. 根据权利要求18所述的雷达系统，其特征在于，所述中央处理器用于对所述第一数据进行测速处理和测角处理，以得到所述第一雷达传感器的测速数据和测角数据，包括：
    所述中央处理器用于获取第二频谱数据，所述第二频谱数据为所述第一频谱数据经过测距处理和测速处理后得到的频谱数据，其中，所述第二频谱数据中包含至少一个第二频谱，所述第二频谱为所述第一频谱经过测距处理和测速处理后所得到的频谱；
    所述中央处理器用于对所述第二频谱数据中的所有第二频谱进行累加；
    在所述第二频谱数据的所有所述第二频谱中，所述中央处理器用于对每个第二频谱同一位置的检测点进行快速傅里叶变换，从而得到每个检测点的测角数据。
  20. 根据权利要求18所述的雷达系统，其特征在于，所述中央处理器用于对所述第一数据进行测速处理和测角处理，以得到所述第一雷达传感器的测速数据和测角数据，包括：
    所述中央处理器用于获取第二频谱数据，所述第二频谱数据为所述第一频谱数据经过测距处理和测速处理后得到的频谱数据，其中，所述第二频谱数据中包含至少一个第二频谱，所述第二频谱为所述第一频谱经过测距处理和测速处理后所得到的频谱；
    所述中央处理器用于对所述第二频谱数据中的所有所述第二频谱进行累加；
    所述中央处理器用于对累加后的所述第二频谱数据进行二维恒虚警率，以从所述检测点中获取第一目标检测点，所述第一目标检测点为所述检测点中的部分检测点；
    在所述第二频谱数据的所有所述第二频谱中，所述中央处理器用于分别对每个第一目标检测点进行快速傅里叶变换，从而得到每个第一目标检测点的测角数据。
  21. 根据权利要求15所述的雷达系统，其特征在于，所述第一射频单元还用于发送第一线性调频信号，所述第一线性调频信号用于进行探测；
    所述第一射频单元用于接收回波数据，包括：
    所述第一射频单元用于接收第二线性调频信号，所述第二线性调频信号为所述第一射频单元发送所述第一线性调频信号之后返回的线性调频信号。
  22. 根据权利要求21所述的雷达系统，其特征在于，
    所述第一微处理器用于对所述回波数据进行预处理，以得到第一数据，包括：
    所述第一微处理器用于对所述第二线性调频信号进行预处理，以得到测距数据；
    所述中央处理器用于对所述第一数据进行后处理，以得到所述第一雷达传感器的第二数据，包括：
    所述中央处理器用于，当所述测距数据所对应的第二数字线性调频信号的数据量累积达到一帧时，对所述第二数字线性调频信号进行测速处理和测角处理。
  23. 根据权利要求21所述的雷达系统，其特征在于，所述第一微处理器用于对所述回波数据进行预处理，以得到第一数据，包括：
    所述第一微处理器用于对所述第一射频单元所接收的X个所述第二数字线性调频信号进行下采样，得到Y个第三数字线性调频信号，其中，X和Y均为大于1的正整数，其中，Y小于X；
    所述中央处理器用于对所述第一数据进后处理，以得到所述第一雷达传感器的第二数据，包括：
    所述中央处理器用于对所述第一数据中的第三数字线性调频信号进行测距处理、测速处理和测角处理，以得到测距数据、测速数据和测角数据。
  24. 根据权利要求15至23任一所述的雷达系统，其特征在于，所述第一微处理器用于对所述回波数据进行预处理，以得到第一数据，包括；
    所述第一微处理器用于对所述回波数据进行压缩，以得到第一压缩数据；
    所述中央处理器用于对所述合并数据进行后处理，包括：
    所述中央处理器用于对所述压缩合并数据进行解压缩。
PCT/CN2019/092084 2019-06-20 2019-06-20 一种雷达系统 WO2020252743A1 (zh)

Priority Applications (4)

Application Number Priority Date Filing Date Title
PCT/CN2019/092084 WO2020252743A1 (zh) 2019-06-20 2019-06-20 一种雷达系统
EP19933643.9A EP3968054A4 (en) 2019-06-20 2019-06-20 RADAR SYSTEM
CN201980066312.3A CN112805591A (zh) 2019-06-20 2019-06-20 一种雷达系统
US17/554,605 US20220107390A1 (en) 2019-06-20 2021-12-17 Radar System

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
PCT/CN2019/092084 WO2020252743A1 (zh) 2019-06-20 2019-06-20 一种雷达系统

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US17/554,605 Continuation US20220107390A1 (en) 2019-06-20 2021-12-17 Radar System

Publications (1)

Publication Number Publication Date
WO2020252743A1 true WO2020252743A1 (zh) 2020-12-24

Family

ID=74037609

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/092084 WO2020252743A1 (zh) 2019-06-20 2019-06-20 一种雷达系统

Country Status (4)

Country Link
US (1) US20220107390A1 (zh)
EP (1) EP3968054A4 (zh)
CN (1) CN112805591A (zh)
WO (1) WO2020252743A1 (zh)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11340346B1 (en) 2020-02-10 2022-05-24 Perceptive Inc. Centralized object detection sensor network system
US20230296756A1 (en) * 2022-03-18 2023-09-21 Nvidia Corporation Sensor data based map creation and localization for autonomous systems and applications

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11683049B2 (en) * 2020-02-18 2023-06-20 Infineon Technologies Ag Hardware decompressor for padding non-equidistant data
CN115665891B (zh) * 2022-12-28 2023-03-10 中国电子科技集团公司信息科学研究院 一种去中心化的分布式雷达系统
CN116299474B (zh) * 2023-05-23 2023-09-12 禾多科技（北京）有限公司 一体化雷达装置和车辆避障方法

Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104459669A (zh) * 2014-12-10 2015-03-25 珠海纳睿达科技有限公司 雷达反射信号处理装置及其处理方法
CN108020837A (zh) * 2016-11-01 2018-05-11 北京行易道科技有限公司 雷达、雷达成像的方法、装置和无人驾驶汽车
WO2018089082A1 (en) * 2016-11-09 2018-05-17 Raytheon Company Systems and methods for direction finding using compressive sensing
CN108170373A (zh) * 2017-12-19 2018-06-15 北京云知声信息技术有限公司 一种数据缓存方法、装置及数据传输系统
CN207636770U (zh) * 2017-12-29 2018-07-20 深圳承泰科技有限公司 一种分布式汽车雷达系统、汽车控制器及汽车
CN109001688A (zh) * 2018-05-28 2018-12-14 中国电子科技集团公司第二十九研究所 一种基于雷达信号并行处理的中间数据存储方法及装置
CN109444819A (zh) * 2018-11-29 2019-03-08 加特兰微电子科技(上海)有限公司 雷达系统及其控制方法
CN109683155A (zh) * 2018-12-19 2019-04-26 深圳市易成自动驾驶技术有限公司 传感器融合系统、方法、终端及存储介质

Family Cites Families (17)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US5412414A (en) * 1988-04-08 1995-05-02 Martin Marietta Corporation Self monitoring/calibrating phased array radar and an interchangeable, adjustable transmit/receive sub-assembly
US7085637B2 (en) * 1997-10-22 2006-08-01 Intelligent Technologies International, Inc. Method and system for controlling a vehicle
EP1627237A2 (en) * 2003-05-27 2006-02-22 Clean Earth Technologies, LLC Radar or sensor system with hierarchical architecture and reconfigurable functionality
WO2005069905A2 (en) * 2004-01-16 2005-08-04 Ghz Tr Corporation Methods and apparatus for automotive radar sensors
US7791528B2 (en) * 2008-11-24 2010-09-07 Autoliv Asp, Inc. Method and apparatus for radar signal processing
EP2881752B1 (en) * 2013-12-03 2017-05-10 Nxp B.V. A multichip automotive radar system, a radar chip for such as system, and a method of operating such a system
US10613213B2 (en) * 2016-05-13 2020-04-07 Google Llc Systems, methods, and devices for utilizing radar with smart devices
US20180113209A1 (en) * 2016-10-21 2018-04-26 Waymo Llc Radar generated occupancy grid for autonomous vehicle perception and planning
CN106814353A (zh) * 2017-01-24 2017-06-09 成都泰格微电子研究所有限责任公司 一种雷达信号处理系统
DE102017117729A1 (de) * 2017-08-04 2019-02-07 Infineon Technologies Ag Verteiltes Radarssystem
CN108287335A (zh) * 2018-04-19 2018-07-17 广州合智瑞达科技有限公司 一种利用lfmcw雷达的频率调制信号来对多目标进行测距测速的方法
CN109895763A (zh) * 2018-05-17 2019-06-18 华为技术有限公司 基于超声波雷达的泊车车位检测方法和终端
CN109164428B (zh) * 2018-10-15 2020-06-26 华清瑞达（天津）科技有限公司 雷达数字仿真系统及方法
CN109375203A (zh) * 2018-12-11 2019-02-22 中交公局重庆城市建设发展有限公司 基于fmcw雷达信号处理的测距设备及其测距算法
CN109633652A (zh) * 2019-01-14 2019-04-16 长沙莫之比智能科技有限公司 基于毫米波雷达的机器人避障系统及其应用方法
CN109787759B (zh) * 2019-01-23 2021-10-15 郑州云海信息技术有限公司 一种数据传输方法、系统、装置及计算机可读存储介质
CN109884638A (zh) * 2019-03-13 2019-06-14 苏州理工雷科传感技术有限公司 一种信号处理机、雷达系统及信号处理方法

Patent Citations (8)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN104459669A (zh) * 2014-12-10 2015-03-25 珠海纳睿达科技有限公司 雷达反射信号处理装置及其处理方法
CN108020837A (zh) * 2016-11-01 2018-05-11 北京行易道科技有限公司 雷达、雷达成像的方法、装置和无人驾驶汽车
WO2018089082A1 (en) * 2016-11-09 2018-05-17 Raytheon Company Systems and methods for direction finding using compressive sensing
CN108170373A (zh) * 2017-12-19 2018-06-15 北京云知声信息技术有限公司 一种数据缓存方法、装置及数据传输系统
CN207636770U (zh) * 2017-12-29 2018-07-20 深圳承泰科技有限公司 一种分布式汽车雷达系统、汽车控制器及汽车
CN109001688A (zh) * 2018-05-28 2018-12-14 中国电子科技集团公司第二十九研究所 一种基于雷达信号并行处理的中间数据存储方法及装置
CN109444819A (zh) * 2018-11-29 2019-03-08 加特兰微电子科技(上海)有限公司 雷达系统及其控制方法
CN109683155A (zh) * 2018-12-19 2019-04-26 深圳市易成自动驾驶技术有限公司 传感器融合系统、方法、终端及存储介质

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
See also references of EP3968054A4 *

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11340346B1 (en) 2020-02-10 2022-05-24 Perceptive Inc. Centralized object detection sensor network system
US11860269B2 (en) 2020-02-10 2024-01-02 Perceptive Inc. Centralized object detection sensor network system
US20230296756A1 (en) * 2022-03-18 2023-09-21 Nvidia Corporation Sensor data based map creation and localization for autonomous systems and applications

Also Published As

Publication number Publication date
EP3968054A1 (en) 2022-03-16
CN112805591A (zh) 2021-05-14
US20220107390A1 (en) 2022-04-07
EP3968054A4 (en) 2022-05-11

Similar Documents

Publication Publication Date Title
WO2020252743A1 (zh) 一种雷达系统
CN110133650B (zh) 一种基于步进频率合成孔径雷达的近距离rcs测量电子系统
CN107810430A (zh) 用于停车辅助的毫米波传感器系统
CN108604733B (zh) 用于手势和动作分析的毫米波传感器系统
CN109655831B (zh) 一种无人机sar成像方法及装置
CN111257655B (zh) 射频传感器被截获距离测试设备
CN113376184B (zh) 一种土壤湿度检测方法及装置
CN104777467B (zh) 基于频率扫描天线的目标检测方法
CN108597234A (zh) 一种基于高分辨率雷达的智能交通检测仪
US10281575B2 (en) Method, system and device for radar based high resolution object detection
CN101852793A (zh) 一种射频无线湿度传感系统
WO2023245862A1 (zh) 一种mimo雷达监测系统及基于mimo雷达监测系统的监测方法
CN213904292U (zh) 手势识别系统
WO2022266817A1 (zh) 天线阵列、探测设备和终端
CN111443336B (zh) 一种降低fmcw雷达系统数据传输吞吐量的方法
CN110944380A (zh) 一种面向毫米波通信的mimo并行通道探测装置及其方法
CN110554440A (zh) 星载微波辐射测量系统及测量方法
CN101969330A (zh) 一种基于软件无线电的宽频段波达方向识别设备
CN113093168A (zh) 测距测速方法、装置、雷达及可读存储介质
CN113490949A (zh) 一种目标的特征提取方法及装置
CN209496136U (zh) 基于mimo的毫米波雷达有轨电车防碰撞预警系统
CN217846611U (zh) 雷达传感器及电子设备
CN105223556B (zh) L型收发阵列天线前端及其信号处理方法
CN111464215B (zh) 一种信号采集与处理系统及方法
CN106291527A (zh) 基于固定指向天线宽波束测角的雷达装置

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: 19933643

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2019933643

Country of ref document: EP

Effective date: 20211207

NENP Non-entry into the national phase

Ref country code: DE