阅读说明：本技术 一种雷达探测方法及相关装置 (Radar detection method and related device ) 是由 李孟麟 李强 姜彤 于 2020-06-30 设计创作，主要内容包括：本申请实施例提供一种雷达探测方法及相关装置,该方法包括：按照时域滑动步长从雷达的拍频信号中截取多个测量单元MU,其中,所述多个测量单元MU中的每个MU的时域长度均大于所述预设滑动步进的时域长度；确定所述每个MU的频率信息,根据所述每个MU的频率信息分别得到一个雷达点云探测结果,其中所述探测结果包括目标物的速度或目标物的距离中的至少一项。采用本申请实施例,能够在不明显提高成本的情况下,提高雷达出点率(即点云密度)且不丢失信噪比。(The embodiment of the application provides a radar detection method and a related device, wherein the method comprises the following steps: intercepting a plurality of measurement units MU from a beat signal of the radar according to a time domain sliding step length, wherein the time domain length of each MU in the plurality of measurement units MU is larger than the time domain length of the preset sliding step; and determining the frequency information of each MU, and respectively obtaining a radar point cloud detection result according to the frequency information of each MU, wherein the detection result comprises at least one of the speed of the target object or the distance of the target object. By adopting the embodiment of the application, the point emergence rate (namely point cloud density) of the radar can be improved without obviously increasing the cost, and the signal to noise ratio is not lost.)

1. A radar detection method, comprising:

intercepting a plurality of Measurement Units (MU) from a beat signal of the radar according to a time domain sliding step length, wherein the time domain length of each MU in the MU is larger than the time domain length of the preset sliding step;

determining frequency information of each MU;

and respectively obtaining a radar point cloud detection result according to the frequency information of each MU, wherein the detection result comprises at least one of the speed of the target object or the distance of the target object.

2. The method of claim 1,

the signal energy accumulated in the time domain length of a first MU of the plurality of MUs can ensure that the signal-to-noise ratio of the first MU is higher than a preset threshold.

3. The method according to claim 1 or 2, characterized in that:

a time domain length of a second MU of the plurality of MUs is the same as a time domain length of a third MU of the plurality of MUs, or,

the time domain length of the second MU is different from the time domain length of the third MU.

4. The method of any of claims 1-3, wherein the beat signal of the radar comprises a first slope beat signal obtained by a first laser transceiver and a second slope beat signal obtained by a second laser transceiver;

intercepting a plurality of measurement units MU from a beat signal of the radar according to a time domain sliding step length comprises:

intercepting a plurality of first measurement units MU from the beat signal of the first slope according to a time domain sliding step size, and intercepting a plurality of second measurement units MU from the beat signal of the second slope according to the time domain sliding step size;

wherein one MU comprises one first MU and one second MU, the one first MU and the one second MU being synchronized in time domain.

5. The method of claim 4,

the first slope is a positive slope, and the second slope is a negative slope;

alternatively, the first and second electrodes may be,

the first slope is a positive slope or a negative slope, and the second slope is a zero slope.

6. The method of any of claims 1-3, wherein the beat signal of the radar comprises a positive slope beat signal and a negative slope beat signal alternating in time domain by a third laser transceiver; each measurement unit MU of the plurality of measurement units MU includes a positive slope portion and a negative slope portion.

7. A signal processing apparatus, characterized by comprising:

the acquisition unit is used for acquiring a plurality of Measurement Units (MU) from a beat signal of the radar according to a time domain sliding step length, wherein the time domain length of each MU in the MU is greater than the time domain length of the preset sliding step length;

a determining unit configured to determine frequency information of each MU;

and the analysis unit is used for respectively obtaining a radar point cloud detection result according to the frequency information of each MU, wherein the detection result comprises at least one of the speed of the target object or the distance of the target object.

8. The apparatus of claim 7, wherein the signal energy accumulated in the time domain length of a first MU of the plurality of MUs ensures that the signal-to-noise ratio of the first MU is higher than a predetermined threshold.

9. The apparatus of claim 7 or 8, wherein:

a time domain length of a second MU of the plurality of MUs is the same as a time domain length of a third MU of the plurality of MUs, or,

the time domain length of the second MU is different from the time domain length of the third MU.

10. The apparatus of any of claims 7-9, wherein the beat signal of the radar comprises a first slope beat signal obtained by a first laser transceiver and a second slope beat signal obtained by a second laser transceiver;

the intercepting unit is specifically configured to intercept a plurality of first measurement units MU from the first slope beat signal according to a time domain sliding step length, and intercept a plurality of second measurement units MU from the second slope beat signal according to the time domain sliding step length;

wherein one MU comprises one first MU and one second MU, the one first MU and the one second MU being synchronized in time domain.

11. The apparatus of claim 10,

the first slope is a positive slope, and the second slope is a negative slope;

alternatively, the first and second electrodes may be,

the first slope is a positive slope or a negative slope, and the second slope is a zero slope.

12. The apparatus of any of claims 7-9, wherein the beat signal of the radar comprises a positive slope beat signal and a negative slope beat signal alternating in time domain by a third laser transceiver; each measurement unit MU of the plurality of measurement units MU includes a positive slope portion and a negative slope portion.

13. A processor, characterized in that the processor is configured to invoke a computer program stored in a computer readable storage medium to implement the method of any of claims 1-6.

14. A lidar system comprising a processor, a memory, and a laser, wherein the laser is configured to emit a laser signal, wherein the memory is configured to store a computer program, and wherein the processor is configured to invoke the computer program to perform the method of any of claims 1-6.

15. A computer-readable storage medium, in which a computer program is stored which, when run on a processor, implements the method of any one of claims 1-6.

Technical Field

The present application relates to the field of laser radar technologies, and in particular, to a radar detection method and a related apparatus.

Background

An FMCW radar is a range finding device, which contains different sub-categories, for example, an fm continuous wave radar using radio waves is called FMCW RADAR, and for example, an fm continuous wave radar using laser light is called FMCW LIDAR. In fig. 1, the radar generates a radio frequency or laser signal that is frequency modulated, and divides the generated frequency modulated signal into two paths, wherein one path is used as a local reference signal (also called local oscillator signal), and the other path is emitted to a target object to be measured (also called reflector) and reflected by the surface of the target object to form an echo signal.

Fig. 2 illustrates the processing procedure of the FMCW radar for the reference signal and the echo signal, as shown in fig. 2 (a), and the thick line illustrates the frequency change of the frequency modulation signals of the transmission signal and the reference signal with time, wherein the signal frequency increases from low to high with time in the first half of the time, and the signal frequency decreases from high to low with time in the second half of the time. The thin lines illustrate the echo signals. The echo signal and the reference signal may output a beat signal through the mixer, and the frequency of the beat signal is the frequency difference between the reference signal and the echo signal, as shown in part (b) of fig. 2. Ideally, the beat signal has a fixed frequency (as shown in the dotted line), and as shown in fig. 2 (c), the frequency of the beat signal can be detected by performing frequency domain analysis (generally, FFT) on the beat signal, where the frequency has a one-to-one correspondence relationship with the distance and velocity of the target object, so that the velocity and distance information of the target object can be calculated from the frequency of the beat signal.

As shown in fig. 3, a laser radar generally includes a scanner, the scanner emits laser light in different directions by rotating the laser, and then collects the measurement results in each direction to obtain a 3D point cloud image, and fig. 4 illustrates a 3D point cloud image of the laser radar. The point cloud resolution of the laser radar is an important index, and the high-resolution point cloud is very important for a machine vision algorithm adopted by automatic driving and the like and is very important for realizing functions such as point cloud segmentation, target identification and the like. At present, only one measurement can be completed within the duration of one up-down chirp pair (i.e., the frequency modulation period of one radar), and the resolution of the obtained point cloud (also called point cloud density or point output rate) is low in this case, and a currently common method for improving the resolution of the point cloud is to add more lasers in a laser radar and upgrade 32-line LIDAR (i.e., 32 lasers work simultaneously) to 128-line LIDAR (there are 128 lasers, or 128 lasers are simulated by an optical device).

However, the way of increasing the resolution of the point cloud by stacking hardware significantly increases the cost, volume and power consumption of the lidar.

Disclosure of Invention

The embodiment of the application discloses a radar detection method and a related device, which can improve the point rate (namely point cloud density) of a radar without obviously improving the cost and avoid losing the signal to noise ratio.

In a first aspect, an embodiment of the present application discloses a radar detection method, including:

intercepting a plurality of Measurement Units (MU) from a beat signal of the radar according to a time domain sliding step length, wherein the time domain length of each MU in the MU is larger than the time domain length of the preset sliding step;

determining frequency information of each MU;

and respectively obtaining a radar point cloud detection result according to the frequency information of each MU, wherein the detection result comprises at least one of the speed of the target object or the distance of the target object.

In the method, the measuring unit MU is intercepted from the beat frequency signal in a sliding window mode, and the sliding stepping time domain length in the time domain is smaller than the intercepted time domain length of each MU, so that any two MUs adjacent to each other in the time domain share a part of frequency information, even if more MUs are intercepted, enough signal energy accumulation in each MU can be ensured, and the signal-to-noise ratio is ensured. Therefore, in this way, the radar point rate (i.e. the point cloud density) can be increased without significantly increasing the cost, and the signal-to-noise ratio is not lost.

In a possible implementation manner of the first aspect, the signal energy accumulated in the time domain length of a first MU of the plurality of MUs can ensure that the signal-to-noise ratio of the first MU is higher than a preset threshold.

In one possible implementation form of the first aspect,

a time domain length of a second MU of the plurality of MUs is the same as a time domain length of a third MU of the plurality of MUs, or,

the time domain length of the second MU is different from the time domain length of the third MU.

In one possible implementation manner of the first aspect, the beat signal of the radar includes a first slope beat signal obtained by a first laser transceiver and a second slope beat signal obtained by a second laser transceiver;

intercepting a plurality of measurement units MU from a beat signal of the radar according to a time domain sliding step length comprises:

intercepting a plurality of first measurement units MU from the beat signal of the first slope according to a time domain sliding step size, and intercepting a plurality of second measurement units MU from the beat signal of the second slope according to the time domain sliding step size;

wherein one MU comprises one first MU and one second MU, the one first MU and the one second MU being synchronized in time domain.

Optionally, the first slope is a positive slope, and the second slope is a negative slope; or, the first slope is a positive slope or a negative slope, and the second slope is a zero slope.

In the method, the two paths of laser transmitters are used for synchronously transmitting laser, so that the beat frequency signals with two different slopes can be synchronously obtained, and the time for obtaining the beat frequency signals with two slopes is shortened, so that the speed for calculating the point cloud result in radar detection is increased.

In one possible implementation manner of the first aspect, the beat signal of the radar includes a positive slope beat signal and a negative slope beat signal alternately obtained by a third laser transceiver in a time domain; each measurement unit MU of the plurality of measurement units includes a positive slope portion and a negative slope portion.

In a second aspect, an embodiment of the present application provides a signal processing apparatus, including:

the acquisition unit is used for acquiring a plurality of Measurement Units (MU) from a beat signal of the radar according to a time domain sliding step length, wherein the time domain length of each MU in the MU is greater than the time domain length of the preset sliding step length;

a determining unit configured to determine frequency information of each MU;

and the analysis unit is used for respectively obtaining a radar point cloud detection result according to the frequency information of each MU, wherein the detection result comprises at least one of the speed of the target object or the distance of the target object.

In the method, the measuring unit MU is intercepted from the beat frequency signal in a sliding window mode, and the sliding stepping time domain length in the time domain is smaller than the intercepted time domain length of each MU, so that any two MUs adjacent to each other in the time domain share a part of frequency information, even if more MUs are intercepted, enough signal energy accumulation in each MU can be ensured, and the signal-to-noise ratio is ensured. Therefore, in this way, the radar point rate (i.e. the point cloud density) can be increased without significantly increasing the cost, and the signal-to-noise ratio is not lost.

In a possible implementation manner of the second aspect, the signal energy accumulated in the time domain length of the first MU in the plurality of MUs can ensure that the signal-to-noise ratio of the first MU is higher than a preset threshold.

In one possible implementation of the second aspect,

a time domain length of a second MU of the plurality of MUs is the same as a time domain length of a third MU of the plurality of MUs, or,

the time domain length of the second MU is different from the time domain length of the third MU.

In one possible implementation manner of the second aspect, the beat signal of the radar includes a first slope beat signal obtained by a first laser transceiver and a second slope beat signal obtained by a second laser transceiver;

the intercepting unit is specifically configured to:

intercepting a plurality of first measurement units MU from the beat signal of the first slope according to a time domain sliding step size, and intercepting a plurality of second measurement units MU from the beat signal of the second slope according to the time domain sliding step size;

wherein one MU comprises one first MU and one second MU, the one first MU and the one second MU being synchronized in time domain.

Optionally, the first slope is a positive slope, and the second slope is a negative slope; or, the first slope is a positive slope or a negative slope, and the second slope is a zero slope.

In the method, the two paths of laser transmitters are used for synchronously transmitting laser, so that the beat frequency signals with two different slopes can be synchronously obtained, and the time for obtaining the beat frequency signals with two slopes is shortened, so that the speed for calculating the point cloud result in radar detection is increased.

In one possible implementation manner of the second aspect, the beat signal of the radar includes a positive slope beat signal and a negative slope beat signal alternately obtained by a third laser transceiver in a time domain; each measurement unit MU of the plurality of measurement units MU includes a positive slope portion and a negative slope portion.

In a third aspect, an embodiment of the present application provides a processor, where the processor is configured to invoke a computer program stored in a computer-readable storage medium to implement the method of the first aspect or any possible implementation manner of the first aspect.

In a fourth aspect, an embodiment of the present application provides a lidar system, which includes a processor, a memory, and a laser, where the laser is configured to emit a laser signal, the memory is configured to store a computer program, and the processor is configured to invoke the computer program to implement the method according to the first aspect or any possible implementation manner of the first aspect.

In a fifth aspect, the present application provides a computer-readable storage medium, in which a computer program is stored, and when the computer program runs on a processor, the method is implemented according to the first aspect or any possible implementation manner of the first aspect.

Drawings

Fig. 1 is a schematic diagram of a laser radar provided in an embodiment of the present application;

FIG. 2 is a schematic diagram of a beat signal provided by an embodiment of the present application;

FIG. 3 is a schematic deflection diagram of a laser scanning device according to an embodiment of the present disclosure;

FIG. 4 is a schematic diagram of a point cloud effect provided by an embodiment of the present application;

fig. 5 is a schematic diagram of an architecture of a laser radar system according to an embodiment of the present disclosure;

FIG. 6 is a schematic diagram of a beat signal generated by a triangular wave according to an embodiment of the present disclosure;

FIG. 7 is a schematic diagram of a beat signal generated by a sawtooth wave according to an embodiment of the present disclosure;

FIG. 8 is a schematic diagram illustrating a division of beat times according to an embodiment of the present application;

fig. 9 is a schematic diagram of a triangular chirp coherent signal processing provided by an embodiment of the present application;

FIG. 10 is a schematic diagram illustrating a division of beat times according to an embodiment of the present application;

fig. 11 is a schematic flowchart of a radar detection method according to an embodiment of the present application;

fig. 12 is a schematic diagram illustrating a division of a beat signal according to an embodiment of the present application;

fig. 13 is a schematic structural diagram of a laser radar system according to an embodiment of the present disclosure;

fig. 14 is a schematic diagram illustrating a division of a beat signal according to an embodiment of the present application;

FIG. 15 is a schematic diagram of a beat signal provided by an embodiment of the present application;

fig. 16 is a schematic structural diagram of a laser radar system according to an embodiment of the present disclosure;

fig. 17 is a schematic diagram illustrating a division of a beat signal according to an embodiment of the present application;

fig. 18 is a schematic structural diagram of a signal processing apparatus according to an embodiment of the present application.

Detailed Description

The embodiments of the present application will be described below with reference to the drawings.

Laser radar in the embodiment of the application can be applied to various fields such as intelligent transportation, autopilot, atmospheric environment monitoring, geographical mapping, unmanned aerial vehicle, can accomplish functions such as distance measurement, speed measurement, target tracking, formation of image discernment.

Referring to fig. 5, fig. 5 is a schematic structural diagram of a lidar system according to an embodiment of the present disclosure, where the lidar system is configured to detect information of a target 505, and the lidar system includes:

the Laser 501, which may be a Tunable Laser (TL), is used to generate a Laser signal, which may be a chirped Laser signal, and the modulation waveform of the Laser signal frequency may be a sawtooth wave, a triangular wave, or other waveform.

The splitter device 502 is configured to split laser light generated by the laser 501 to obtain a transmit signal and a Local Oscillator (LO), where the LO is also referred to as a reference signal. Optionally, a collimating lens 500 may be disposed between the laser 501 and the branching device, and the lens 500 is used for beam shaping the laser signal transmitted to the branching device 502.

A collimator 503 for coupling the transmit signal into the scanner 504 with maximum efficiency.

The scanner 504, also called as a 2D scanning mechanism, is configured to emit the emission signal at a certain angle, and after the emission signal is emitted, the emission signal is reflected by the target 505 to form an echo signal; the scanner 504 is then also configured to receive the echo signals, which are combined with the local oscillator signal at a mixer 510 after passing through corresponding optics, such as a mirror 506 (optional) and a receive optic 508 (optional).

And the mixer 510 is configured to perform frequency mixing processing on the local oscillation signal and the echo signal to obtain a beat signal.

A detector 520 for extracting the beat signal from the mixer. For example, the detector 520 may be a balanced detector (BPD).

An Analog Digital Converter (ADC) 511 is used for sampling the beat frequency signal, and this sampling is essentially the process of converting the Analog signal into a digital signal.

And a processor 512, which may include devices with computing power, such as a Digital Signal Processor (DSP), a Central Processing Unit (CPU), an Accelerated Processing Unit (APU), an image processing unit (GPU), a microprocessor or a microcontroller, for example, and which is not shown in the drawings, and is configured to process the sampled beat frequency signal to obtain information, such as speed and distance of the target object.

In the embodiment of the present application, the target 505 is also referred to as a reflector, the target 505 may be any object in the scanning direction of the scanner 504, for example, a person, a mountain, a vehicle, a tree, a bridge, etc., and fig. 5 illustrates a vehicle as an example.

In the embodiment of the present application, the operation of processing the beat signals obtained by sampling to obtain the information of the speed, the distance, and the like of the target object may be performed by one or more processors 512, for example, by one or more DSPs, or may be performed by one or more processors 512 in combination with other devices, for example, by one DSP in combination with one or more Central Processing Units (CPUs). The processing of the beat signal by the processor 512 may be realized by calling a computer program stored in a computer-readable storage medium, which includes but is not limited to a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), or a portable read-only memory (CD-ROM), and may be configured in the processor 512 or independent of the processor 512.

In the embodiment of the present application, some of the above-mentioned devices may be single-part or multiple-part, for example, the laser 501 may be one or multiple, and in the case of one laser 501, the one laser 501 may alternately emit a laser signal with a positive slope and a laser signal with a negative slope in the time domain; when there are two lasers 501, one of which emits a laser signal with a positive slope and the other of which emits a laser signal with a negative slope, the two lasers 501 may emit laser signals synchronously.

As shown in fig. 6, taking the modulation waveform of the laser signal frequency as a triangular wave chirp as an example, the echo signal is mixed with the local oscillator signal LO after a flight time, where the flight time is the time from the emission of the transmit signal divided by the laser signal to the return of the echo signal, and the beat signal generated by the echo signal and the local oscillator signal after the flight time is constant within a certain time, which can accurately reflect the distance and speed information of the target object, and the time is the beat time. The beat signal needs to include a beat frequency f corresponding to a positive slope₁And a beat frequency f corresponding to a negative slope₂Frequency spectrum f related to the velocity of the object_{Speed of rotation}Can be expressed as f_{Speed of rotation}＝(f₁-f₂) /2, frequency f related to distance of target_{Distance between two adjacent plates}Can be expressed as f_{Distance between two adjacent plates}＝(f₁+f₂)/2. To obtain f_{Speed of rotation}And f_{Distance between two adjacent plates}Then the distance of the target object (and the laser radar) and the moving speed of the target object can be calculated.

The coherent lidar needs to measure information of a target object at a longer distance, and needs to increase beat frequency time of each Measurement Unit (MU) (one Measurement Unit MU is used for obtaining a detection result of one Measurement point), so that more energy is accumulated in the process of processing beat frequency signals, and a higher signal-to-noise ratio is achieved. In practical applications, the lidar is required to have a higher out-point rate, so that the lidar obtains higher field resolution and frame rate.

As shown in fig. 7, a commonly used sawtooth linear frequency modulation coherent signal processing method is to use the time left after subtracting the flight time corresponding to the farthest detection distance from the modulation period T as a beat time TB, collect data in the beat time TB, perform fourier transform to obtain the frequency of a real-time beat signal, and further calculate the distance of a target object according to the frequency, where TM of each measurement point is not greater than TB. Furthermore, the speed of the target object can be obtained by matching with other measuring channels with different chirp rates.

In order to improve the point-out rate of the laser radar, a plurality of measuring units MU can be divided within a single beat frequency time, as shown in fig. 8, the beat frequency time is divided into three segments to obtain three MUs (corresponding to three measuring points), and one detection can be completed based on signals of each MU respectively, namely, the point-out rate can be three times that of the original point-out rate, but the beat frequency time of each MU is also changed into 1/3 of the original point-out rate.

As shown in fig. 9, the conventional triangular chirp coherent signal processing method divides the respective flight time and beat frequency time in the positive slope modulation cycle and the negative slope modulation cycle, the signal processing method in the respective beat frequency time is the same as the sawtooth chirp coherent signal processing method, and the upper half cycle (which may also be described as the upper half cycle corresponding to the measurement point) of the measurement unit MU processes to obtain the beat frequency f₁Processing the next half period of the MU (which can also be described as the next half period corresponding to the measurement point) to obtain the beat frequency f₂Therefore, the distance and speed information of the target can be obtained through one measuring channel.

To improve the point-out rate of the lidar, the beat frequency time in the positive slope period and the negative slope period may be divided into a plurality of measurement times, as shown in fig. 10, the beat frequency time T _ τ in the negative slope period is divided into three measurement times T corresponding to three measurement units MU respectively₀Based on the above, each measurement unit MU can complete one detection, i.e. three times of the original point rate, but at the same time, the beat frequency time of each measurement unit MU also becomes 1/3.

It can be seen that the above manner of increasing the point-out rate reduces the beat frequency time of each measurement unit MU, and reduces the energy accumulation of useful signals during time-frequency conversion, thereby affecting the signal-to-noise ratio of measurement and reducing the detection distance of coherent measurement.

In order to guarantee a higher out-point rate and a higher signal-to-noise ratio, the embodiment of the present application provides the method shown in fig. 11.

Referring to fig. 11, fig. 11 is a radar detection method provided by an embodiment of the present application, which may be implemented based on components in the laser radar system shown in fig. 5, and some operations in the following description are performed by a signal processing apparatus, which may be the processor 512 or an apparatus in which the processor 512 is disposed, for example, a laser radar system in which the processor 512 is disposed or a module in the laser radar system, where the method includes, but is not limited to, the following steps:

step S1101: the signal processing device intercepts a plurality of MUs from the beat frequency signal of the radar according to the time domain sliding step length.

The intercepted time domain length of each Measuring Unit (MU) in the MUs is greater than the time domain length of the time domain sliding step length, each measuring Unit MU is used as a measuring signal of one measuring point, and the intercepted time domain length of each MU in the MUs is greater than the time domain length of the time domain sliding step length, so that partial repetition of the beat frequency signals can exist in any two MUs adjacent to each other in the time domain, the utilization rate of the beat frequency signals is improved to the maximum extent, meanwhile, the energy of the signals in a single MU is enhanced, and the signal to noise ratio can be improved.

For example, as shown in fig. 12, in the method for processing a chirp coherent lidar signal modulated according to a sawtooth wave, the scanner 504 drives the transmitted signal to deflect, so that the light beam of the transmitted signal (also called a laser signal) moves on the surface of different target objects, and therefore the target object represented by the reflected echo signal also changes with the deflection angle of the scanner 504. The beat signals formed by the echo signal reflected by the target object and the local oscillator signal LO can be respectively expressed as follows:

the beat signal formed by the echo signal reflected by the target 1 and the local oscillator signal LO is: omega_D1+k*τ₁

Wherein, ω is_D1＝2π*v₁/λ，τ₁2R/C, v1 refers to the velocity of target 1, λ is the laser wavelength, R is the distance of target 1, C is the speed of light, τ₁Is the time, ω, of the reflection of the laser back to the lidar from the transmission to the target 1 echo_D1Is the beat frequency shift caused by the velocity of target 1 and k is the chirp rate of the transmitted laser chirp.

The beat signal formed by the echo signal reflected by the target 2 and the local oscillator signal LO is: omega_D2+k*τ₂

τ₂Is the time, ω, of the reflection of the laser back to the lidar from the transmission to the target 2 echo_D2Is the beat frequency shift caused by the velocity of the target 2.

The beat signal formed by the echo signal reflected by the target 3 and the local oscillator signal LO is: omega_D3+k*τ₃

τ₃Is the time, omega, of the reflection of the laser back to the lidar from the transmission to the target 3_D3Is the beat frequency shift caused by the velocity of the target 3.

When processing the beat frequency signal, intercepting the signal from the beat frequency signal by adopting a segmented intercepting mode, wherein each segment of the signal is taken as a measuring unit MU, and the time length of the intercepted (or sampled) measuring unit MU is represented as T_MAnd after being processed, each intercepted measuring unit MU can output a group of information such as distance, speed, amplitude and the like. In the embodiment of the application, in any two adjacent measurement units MU in the time domain, the distance between the interception start time of the latter measurement unit and the interception start time of the former measurement unit is τ, and τ is less than T_MWhere τ is the previous time-domain sliding step.

Optionally, the time domain length T of the first MU_MThe signal energy accumulated in the first MU can ensure that the signal-to-noise ratio of the first MU is higher than a preset threshold, wherein the first MU is any intercepted segment of MU. That is, T_MThe selection is not random, and is mainly based on the requirement of meeting the signal-to-noise ratio of effective information extraction of a data processing method, if Fourier transform is adopted to extract beat frequency, T is_MThe longer the time, the targetThe more energy is accumulated in the beat frequency, the higher the signal-to-noise ratio is, so the time domain length T of any MU is_MIt needs to be longer than a certain length and cannot be selected at will.

As shown in fig. 12, for the same beat frequency duration, only three measurement units MU (corresponding to three measurement points, i.e., measurement point 1, measurement point 4, and measurement point 7, respectively) can be obtained according to the prior art, and seven measurement units MU (corresponding to seven measurement points, i.e., measurement point 1, measurement point 2, measurement point 3, measurement point 4, measurement point 5, measurement point 6, and measurement point 7, respectively) can be obtained by sampling the above scheme.

There are many possible situations for the time domain length of each measurement unit MU, which are as follows:

optionally, the truncated time domain lengths of the multiple MUs are all the same, that is, if any two MUs are selected, for example, the second MU and the third MU, the time domain length of the second MU is the same as the time domain length of the third MU.

Optionally, the truncated time domain lengths of the plurality of MUs are not exactly the same, that is, if two MUs are taken, such as the second MU and the third MU, the time domain length of the second MU may be different from the time domain length of the third MU.

In the embodiment of the application, the structures of the laser radar transmitting and receiving signals can be set as required, the embodiment of the application exemplifies the following two structures, and explains specific how to capture the MU according to the two different structures.

The first structure, as shown in fig. 13, is a dual-channel lidar system, and the structure shown in fig. 13 can be considered to be obtained by adding corresponding devices on the basis of fig. 5, for example, some devices in fig. 5 are duplicated in fig. 13, such as a detector, a shunt device, and the like. Specifically, the structure shown in fig. 13 includes an ADC, a first laser transceiver, a second laser transceiver, a beam combiner, a collimator, a scanner, and a target, where the first laser transceiver includes a laser 1, a splitter 1, a mixer 1, a receiving mirror 1, and a detector 1; the second laser transceiver comprises a laser 2, a shunt device 2, a mixer 2, a receiving mirror 2 and a detector 2.

In the structure shown in fig. 13, the signal processing flow is as follows:

the laser 1 is first slope modulated and its frequency variation curve can be expressed as ω (t) ═ ω₁+ κ t, which may be referred to as laser 1 frequency, the laser signal generated by the laser 1 is split by the splitter device 1 to obtain the local oscillator signal LO1 and the transmit signal;

the laser 2 is a second slope modulation, and its frequency variation curve can be represented as ω (t) ═ ω₂κ t, which may be referred to as laser 2 frequency, the laser 2 generates a laser signal, and the laser signal is split by the splitter device 2 to obtain a local oscillator signal LO2 and a transmit signal;

optionally, the first laser transceiver and the second laser transceiver may transmit and receive signals synchronously.

The emission signals obtained by the respective branches of the branch device 1 and the branch device 2 are converged in the beam combiner and then emitted to the scanner through the collimator, the scanner emits the emission signals to a target object according to a certain angle, correspondingly, the target object reflects echoes to form echo signals on the scanner, one part of the echo signals (namely echo signals 1) reach the frequency mixer 1 through the receiving mirror 1, and the other part of the echo signals (namely echo signals 2) reach the frequency mixer 2 through the receiving mirror 2;

the frequency mixer 1 carries out frequency mixing on the echo signal 1 and the local oscillation signal LO1 to obtain a beat frequency signal 1, wherein the obtained beat frequency signal 1 is a first slope beat frequency signal;

the frequency mixer 2 performs frequency mixing on the echo signal 2 and the local oscillator signal LO2 to obtain a beat signal 2, wherein the obtained beat signal 2 is a second slope beat signal;

the detector 1 accordingly acquires the beat signal 1 and the detector 2 accordingly acquires the beat signal 2.

As shown in FIG. 13, if the beat frequency of the object 1 is shifted by k τ due to the distance₁The beat frequency shift due to velocity is ω_D1Then, the beat frequency signal 1 of the object 1 acquired by the detector 1 is ω_D1+κτ₁Go forward and go backThe beat frequency signal 2 of the object 1 collected by the detector 2 is-omega_D1+κτ₁. Similarly, if the beat frequency of the target 2 generated by the distance is shifted to k τ₂The beat frequency shift due to velocity is ω_D2Then, the beat frequency signal 1 of the object 1 acquired by the detector 1 is ω_D2+κτ₂(ii) a The beat frequency signal 2 of the target 1 collected by the detector 2 is-omega_D2+κτ₂. In addition, the beat signal corresponding to the target 3 is analogized, and is not described one by one here.

In the structure shown in fig. 13, optionally, the intercepting, according to the time domain sliding step, the multiple measurement units MU from the beat signal of the radar may specifically be: intercepting a plurality of first measurement units MU from the beat signal of the first slope according to a time domain sliding step length, and synchronously intercepting a plurality of second measurement units MU from the beat signal of the second slope according to the time domain sliding step length; wherein one MU comprises one first MU and one second MU, the one first MU and the one second MU being synchronized in time domain. It can be understood that, when viewed in the sequence of the time domains, the time domain of the ith first measurement unit MU is synchronized with the time domain of the ith second measurement unit MU, and i is a positive integer.

As shown in fig. 14, time domain T_M1The obtained beat frequency signal can be called as a measuring unit 1, time domain T_M2The obtained beat frequency signal can be called as a measuring unit 2, time domain T_M3The resulting beat signal may be referred to as a measurement unit 3, where T_M1And T_M2Has a time domain deviation of Δ T from the starting point of (2)₁And T is_M2And T_M3Has a time domain deviation of Δ T from the starting point of (2)₂Then, Δ T₁And Δ T₂Can be regarded as the time domain sliding step length, Delta T₁And Δ T₂Are all less than T_M1Time domain length of (T)_M2Time domain length of (T)_M3Time domain length of (a). In addition, in the time domain T_M1In practice, the measuring unit 1 comprises a first measuring unit MU and a second measuring unit MU, and the measuring unit 1 is used as the measuring signal of the measuring point 1; in the time domain T_M2The measuring unit 2 actually comprises a first measuring unit MU and aA second measuring unit MU, the measuring unit 2 being adapted to act as a measuring signal for the measuring point 2; in the time domain T_M3The measuring unit 3 also comprises a first measuring unit MU and a second measuring unit MU, the measuring unit 3 being intended to serve as a measuring signal for the measuring point 3.

Two lines of beat signals detected by the detector 1 and the detector 2 are collected (or sampled, generally, digital signals are sampled from analog signals) through an ADC (analog to digital converter), and are synchronously processed, wherein after the flight time, the time domain length is T_M1The two paths of signals are used as measuring signals of a measuring point 1; at a time Δ T after the start of measurement Point 1₁The time is the starting point and the time domain length is T_M2The two paths of signals are used as measuring signals of a measuring point 2; at time Δ T after the start of measurement Point 2₂Time as starting point and time domain length as T_M3As the measurement signal of the measurement point 3.

Here again, the time-domain length of the individual measurement units MU, such as T_M1、T_M2And T_M3Which may or may not be equal, each time domain sliding step, e.g., Δ T₁And Δ T₂May or may not be equal. One of the criteria for determining the time domain length is the snr of the measurement unit MU obtained after the truncation, for example, a threshold may be set, and only when the power spectrum amplitude of a certain frequency component in a certain period of time is higher than the average power of the surrounding noise by the threshold, the frequency component is determined to be the useful signal frequency, so that the frequency domain signal in the certain period of time is truncated as a measurement unit MU.

When the power spectral density is obtained according to each measurement unit MU, as shown in fig. 15, the frequency of the useful signal including information such as the target distance and the speed can be distinguished from the frequency of the useful signal, and the longer the time domain of each measurement unit is, the higher the spectral amplitude of the useful signal is, the more the spectral peak can be extracted, so that the detection accuracy of the coherent measurement can be improved.

Optionally, the distance is R₁The echo signal and the distance R reflected by the target 1 of₂When the echo signals reflected by the target 2 reach the similar signal power spectrum amplitude, the time domain length T of the measuring unit corresponding to the target 1_M1Corresponding number of sampling points N₁Time domain length T of the measurement unit corresponding to target 2_M2The number N of corresponding sampling points (i.e. sampling points collected by ADC)₂The following formula should be satisfied:

optionally, the first slope is a positive slope, and the second slope is a negative slope; or, the first slope is a positive slope or a negative slope, and the second slope is a zero slope.

A second configuration, as shown in fig. 16, is a single channel lidar system, which has one less laser transceiver than the configuration shown in fig. 13, and the configuration shown in fig. 16 includes one laser transceiver, which is denoted as a third laser transceiver; the third laser transceiver includes a laser, a detector, a mixer, a branching device, and a receiving mirror, the structure of fig. 16 is equivalent to the structure of fig. 5 with one more receiving mirror, and the receiving mirror is disposed between the scanner and the BPD, and the basic operation principle of the structure of fig. 16 can refer to the foregoing description of the structure of fig. 5. Because the structure shown in fig. 16 adopts a single-channel laser radar system, the laser can emit triangular wave linear frequency modulation laser, the first half cycle is positive slope modulation, and the second half cycle is negative slope modulation, so as to obtain a frequency signal with a positive slope and a frequency signal with a negative slope; thus, the detector can subsequently acquire the beat frequency signal with positive slope and the beat frequency signal with negative slope which are alternated on the time domain; in this case each of the plurality of measurement units MU taken in time domain sliding steps comprises a positive slope portion and a negative slope portion.

As shown in fig. 17, time domain T_M1The obtained beat frequency signal can be called as a measuring unit 1, time domain T_M2The obtained beat frequency signal can be called as a measuring unit 2, time domain T_M3The resulting beat signal may be referred to as a measurement unit 3, where T_M1And T_M2Has a time domain deviation of Δ T from the starting point of (2)₁And T is_M2And T_M3Has a time domain deviation of Δ T from the starting point of (2)₂Then, Δ T₁And Δ T₂Can be regarded as the time domain sliding step length, Delta T₁And Δ T₂Are all less than T_M1Time domain length of (T)_M2Time domain length of (T)_M3Time domain length of (a). In addition, in the time domain T_M1In the above, the measurement unit 1 actually includes a positive slope beat frequency signal and a negative slope beat frequency signal staggered in a time domain, and the measurement unit 1 is used as a measurement signal of a measurement point 1; in the time domain T_M2The measuring unit 2 actually comprises a positive slope beat frequency signal and a negative slope beat frequency signal which are staggered in a time domain, and the measuring unit 2 is used as a measuring signal of the measuring point 2; in the time domain T_M3In practice, the measurement unit 3 includes a time-domain staggered positive slope beat frequency signal and a time-domain staggered negative slope beat frequency signal, and the measurement unit 3 is used as a measurement signal of the measurement point 3.

Collecting and processing beat signals detected by the ADC detector, wherein the time domain length is T after the flight time_M1One path of signal is used as a measuring signal of a measuring point 1; at a time Δ T after the start of measurement Point 1₁The time is the starting point and the time domain length is T_M2One path of signal is taken as a measuring signal of the measuring point 2; at time Δ T after the start of measurement Point 2₂Time as starting point and time domain length as T_M3One signal is taken as a measuring signal of the measuring point 3.

Here again, the time-domain length of the individual measurement units MU, such as T_M1、T_M2And T_M3Which may or may not be equal, each time domain sliding step, e.g., Δ T₁And Δ T₂May or may not be equal. One of the criteria for determining the time domain length is the snr of the measurement unit MU obtained after the truncation, for example, a threshold may be set, and only when the power spectrum amplitude of a certain frequency component in a certain period of time is higher than the average power of the surrounding noise by the threshold, the frequency component is determined to be the useful signal frequency, so that the frequency domain signal in the certain period of time is truncated as a measurement unit MU.

Optionally, the time domain length TM of each measurement unit should be greater than the triangular wave modulation period, and the interval between the time domain start times of each measurement unit should be smaller than the triangular wave modulation period.

When the power spectral density is obtained according to each measurement unit MU, as shown in fig. 15, the frequency of the useful signal including information such as the target distance and the speed can be distinguished from the frequency of the useful signal, and the longer the time domain of each measurement unit is, the higher the spectral amplitude of the useful signal is, the more the spectral peak can be extracted, so that the detection accuracy of the coherent measurement can be improved.

Optionally, the distance is R₁The echo signal and the distance R reflected by the target 1 of₂When the echo signals reflected by the target 2 reach the similar signal power spectrum amplitude, the time domain length T of the measuring unit corresponding to the target 1_M1Corresponding number of sampling points N₁Time domain length T of the measurement unit corresponding to target 2_M2Corresponding number of sampling points N₂The following formula should be satisfied:

step S1102: the signal processing means determines frequency information of said each MU.

Specifically, frequency information in the MU can be extracted by performing fourier transform on the MU; the frequency information of each MU can be extracted by performing Fourier transform on each MU of the MUs. Of course, the frequency information of each MU may be extracted in other manners.

Step S1103: and the signal processing device respectively obtains a radar point cloud detection result according to the frequency information of each MU.

It can be understood that the frequency information of each MU can obtain a detection result of one detection point, and one detection point corresponds to one point cloud, so that the frequency information of each MU can be expressed as a radar point cloud detection result.

Wherein the detection result comprises one or more of a speed of the target object or a distance of the target object.

In view of the above-mentioned structure one,

if the first slope is a positive slope or a negative slope, the second slope is a zero slope; then, for the second measurement unit intercepted by the beat frequency signal with zero slope, the frequency change f caused by the speed of the target object can be obtained_{Speed of rotation}The frequency change f caused by the distance of the target object can be obtained by the second measuring unit obtained by intercepting the beat frequency signal with zero slope and the first measuring unit obtained by intercepting the beat frequency signal with positive slope or negative slope_{Distance between two adjacent plates}Thus, based on f_{Speed of rotation}And f_{Distance between two adjacent plates}The distance and the speed of the target object can be obtained respectively.

In view of the above-mentioned structure one,

if the first slope is a positive slope, the second slope is a negative slope; then, a power spectral density function can be determined for the positive slope measurement unit of each measurement unit, and spectral lines above a predetermined signal-to-noise threshold can be extracted as the useful frequency signal f₁And, the negative slope measuring unit of each measuring unit is taken the power spectral density function, and the spectral line higher than the preset signal-to-noise threshold is taken as the useful frequency signal f₂。

For the second structure, a power spectral density function can be obtained for the positive slope part (i.e. the beat signal within the positive slope half period beat time) of each measuring unit, and a spectral line higher than a preset signal-to-noise ratio threshold value is extracted as a useful frequency signal f₁And, the power spectral density function is obtained for the negative slope part (i.e. the beat signal within the negative slope half period beat time) of each measuring unit, and the spectral line higher than the preset signal-to-noise ratio threshold is extracted as the useful frequency signal f₂。

Obtaining a useful signal f corresponding to a positive slope beat frequency₁And a useful signal f corresponding to a negative slope₂Then, the frequencies f for finding the distance to the target object can be obtained according to the following formulas_{Distance between two adjacent plates}And a frequency f for determining the velocity of the object_{Speed of rotation}。

f_{Speed of rotation}＝(f₁-f₂)/2

f_{Distance between two adjacent plates}＝(f₁+f₂)/2

To obtainf_{Speed of rotation}And f_{Distance between two adjacent plates}Then according to f_{Speed of rotation}Calculating the moving speed of the target (relative to the laser radar), and calculating according to f_{Distance between two adjacent plates}And calculating to obtain the distance between the target object (and the laser radar).

It is understood that one probing result can be obtained through step S1103 for each intercepted measurement unit MU, and therefore, after the same step is performed for each intercepted measurement unit MU, a plurality of probing results can be obtained.

In the method shown in fig. 11, the measurement unit MU is intercepted from the beat signal by sliding the sliding window, and since the sliding step time domain length in the time domain is smaller than the intercepted time domain length of each MU, any two MUs adjacent in the time domain share a part of frequency information, so that even if more MUs are intercepted, sufficient signal energy accumulation in each MU can be ensured, and the signal-to-noise ratio is ensured. Therefore, in this way, the radar point rate (i.e. the point cloud density) can be increased without significantly increasing the cost, and the signal-to-noise ratio is not lost.

The method of the embodiments of the present application is set forth above in detail and the apparatus of the embodiments of the present application is provided below.

Referring to fig. 18, fig. 18 is a schematic structural diagram of a signal processing apparatus 180 according to an embodiment of the present disclosure, where the apparatus may be the lidar system, or a processor in the lidar system, or a relevant device disposed in the lidar system. The signal processing device 180 may comprise an intercepting unit 1801, a determining unit 1802 and an analyzing unit 1803, wherein the details of each unit are described below.

An intercepting unit 1801, configured to intercept multiple measurement units MU from a beat signal of a radar according to a time domain sliding step length, where a time domain length of each MU in the multiple MUs is greater than a time domain length of the preset sliding step length;

a determining unit 1802 configured to determine frequency information of the each MU;

an analyzing unit 1803, configured to obtain a radar point cloud detection result according to the frequency information of each MU, where the detection result includes at least one of a speed of the target object or a distance of the target object.

In the method, the measuring unit MU is intercepted from the beat frequency signal in a sliding window mode, and the sliding stepping time domain length in the time domain is smaller than the intercepted time domain length of each MU, so that any two MUs adjacent to each other in the time domain share a part of frequency information, even if more MUs are intercepted, enough signal energy accumulation in each MU can be ensured, and the signal-to-noise ratio is ensured. Therefore, in this way, the radar point rate (i.e. the point cloud density) can be increased without significantly increasing the cost, and the signal-to-noise ratio is not lost.

Optionally, the signal energy accumulated in the time domain length of the first MU in the multiple MUs may ensure that the signal-to-noise ratio of the first MU is higher than a preset threshold.

Optionally, a time domain length of a second MU of the multiple MUs is the same as a time domain length of a third MU of the multiple MUs, or the time domain length of the second MU is different from the time domain length of the third MU.

Optionally, the beat signal of the radar includes a first slope beat signal obtained by a first laser transceiver and a second slope beat signal obtained by a second laser transceiver;

the intercepting unit is specifically configured to:

intercepting a plurality of first measurement units MU from the beat signal of the first slope according to a time domain sliding step length, and synchronously intercepting a plurality of second measurement units MU from the beat signal of the second slope according to the time domain sliding step length; wherein one MU comprises one first MU and one second MU, the one first MU and the one second MU being synchronized in time domain.

Optionally, the first slope is a positive slope, and the second slope is a negative slope; or, the first slope is a positive slope or a negative slope, and the second slope is a zero slope.

In the method, the two paths of laser transmitters are used for synchronously transmitting laser, so that the beat frequency signals with two different slopes can be synchronously obtained, and the time for obtaining the beat frequency signals with two slopes is shortened, so that the speed for calculating the point cloud result in radar detection is increased.

Optionally, the beat signal of the radar includes a positive slope beat signal and a negative slope beat signal alternately obtained by a third laser transceiver in a time domain; each measurement unit MU of the plurality of measurement units MU includes a positive slope portion and a negative slope portion.

It should be noted that the implementation of each unit may also correspond to the corresponding description of the method embodiment shown in fig. 11.

The embodiment of the present application further provides a chip system, where the chip system includes at least one processor, a memory and an interface circuit, where the memory, the interface circuit and the at least one processor are interconnected by a line, and the at least one memory stores instructions; the instructions, when executed by the processor, implement the method flow shown in fig. 11.

Embodiments of the present application also provide a computer-readable storage medium, which stores instructions that, when executed on a processor, implement the method flow illustrated in fig. 11.

Embodiments of the present application also provide a computer program product, which when executed on a processor implements the method flow illustrated in fig. 11.

In summary, by implementing the embodiment of the present application, the measurement unit MU is intercepted from the beat signal by a sliding window, and since the sliding step time domain length in the time domain is smaller than the intercepted time domain length of each MU, two MUs adjacent to each other in any two time domains share a part of frequency information, even if more MUs are intercepted, it can be ensured that enough signal energy is accumulated in each MU, and the signal-to-noise ratio is ensured. Therefore, in this way, the radar point rate (i.e. the point cloud density) can be increased without significantly increasing the cost, and the signal-to-noise ratio is not lost.

One of ordinary skill in the art will appreciate that all or part of the processes in the methods of the above embodiments may be implemented by hardware related to instructions of a computer program, which may be stored in a computer-readable storage medium, and when executed, may include the processes of the above method embodiments. And the aforementioned storage medium includes: various media capable of storing program codes, such as ROM or RAM, magnetic or optical disks, etc.