Ranging method and system based on laser radar saturation signal

Abstract

The invention relates to the field of laser radar ranging methods, in particular to a ranging method and system based on laser radar saturation signals. The method comprises the following steps of: calculating an average value and a standard deviation of a background noise signal by taking a part without an obvious activity signal between a main wave signal and an echo signal in a full-waveform signal of the hyperspectral laser radar as the background noise signal of a corresponding wave band, and finishing subsequent signal filtering and removing a low-intensity signal by using a noise threshold and a low-signal threshold; echo signals in the preprocessed signals are classified, and waveform arrival time is calculated by adopting different time discrimination algorithms based on a classification result; according to the emission time of the main wave signal of each wave band and the waveform arrival time and the light velocity of the echo signal, the calculation of the target distance is realized; the problem that the existing TOF distance measurement method cannot adapt to time errors caused by various waveform distortions when measuring the distance of a high-reflection target, so that the problem of saturation signal distance measurement precision is solved.

Description

Technical Field

[0001] The present invention relates to the field of laser radar ranging methods, and in particular to a ranging method and system based on a laser radar saturation signal. Background Art

[0002] Laser ranging technology is one of the key technologies of laser radar environmental perception. According to the distance measurement principle, it is divided into pulse method, phase method and triangulation method. Pulse ranging method is also called TOF ranging method. TOF ranging method uses the time difference algorithm to calculate the time interval between signal transmission and return to determine the target distance. However, when measuring the distance of highly reflective targets such as smooth metal surfaces and retroreflective targets, the intensity of the echo signal often exceeds the receiving range of the laser radar based on TOF measurement, resulting in saturation distortion of the echo waveform. The accuracy of TOF ranging method, which is based on single-wavelength signal ranging, depends on the complete pulse waveform and the appropriate waveform processing algorithm. However, the fixed waveform moment identification method cannot adapt to the time error caused by various waveform distortions, thereby reducing the accuracy of distance measurement. Summary of the invention

[0003] In view of the shortcomings of the prior art, the present invention provides a distance measurement method and system based on a laser radar saturation signal, which solves the problem that the current TOF distance measurement method cannot adapt to the time error caused by various waveform distortions when measuring the distance of highly reflective targets, thereby reducing the distance measurement accuracy.

[0004] To achieve the above object, the present invention provides the following technical solutions:

[0005] A distance measurement method based on a laser radar saturation signal comprises the following steps:

[0006] S1. Collecting full waveform data containing multiple bands based on a hyperspectral laser radar, wherein each band of the full waveform data contains a main wave signal emitted by the laser and an echo signal reflected by a target;

[0007] S2, obtaining the noise signal between the end time of the main wave signal and the start time of the echo signal in each band to calculate its average value and the sum of several times of the standard deviation to obtain the noise threshold and the low signal threshold, and filtering the waveform data and removing the low intensity band to obtain the non-low intensity signal;

[0008] S3, classifying each echo waveform of the non-low-intensity signal into a saturated waveform with obvious waveform truncation, a normal waveform with obvious peaks, and a U-shaped waveform with distortion between the rising edge and the falling edge;

[0009] S4, respectively calculating the waveform arrival time of the normal waveform, the saturated waveform and the U-shaped waveform by using the peak method, the waveform fitting method and the centroid method;

[0010] S5. Calculate the target distance between the target and the signal receiving position according to the waveform arrival time of the echo waveform, the emission time of the main wave signal and the speed of light.

[0011] Preferably, in step S2, the following steps are specifically included:

[0012] S21, obtaining the ending time of the falling edge of the main wave signal in each band and the starting time of the rising edge of the echo signal in the corresponding band;

[0013] S22, acquiring waveform data between the end time of the falling edge of the main wave signal and the start time of the rising edge of the echo signal in the corresponding band in the waveform data, and marking it as background noise;

[0014] S23, calculating the average value of the noise signal of the background noise; the calculation formula of the noise signal average value is:

[0015]

[0016] In the above formula, μ noise Indicates the average value of the noise signal, wavenoise i represents the noise signal strength at the sampling time corresponding to the i-th sampling point, and n represents a total of n sampling points;

[0017] S24, calculating the standard deviation of background noise; the calculation formula of the standard deviation of background noise is:

[0018]

[0019] In the above formula, σ noise represents the standard deviation of background noise;

[0020] S25, obtaining a noise threshold and a low signal threshold according to the average value of the background noise and the sum of several times the standard deviation;

[0021] The noise threshold is calculated as:

[0022] Noise threshold =μ noise +w Noise σ noise

[0023] The calculation formula for the low signal threshold is:

[0024] LowSignal threshold =μ noise +w LowSignal σ noise

[0025] In the above formula, Noise threshold represents the noise threshold, w NoiseRepresents the noise factor, LowSignal threshold Indicates the low signal threshold, w LowSignal Indicates low signal coefficient;

[0026] S26, filtering out noise in the waveform data based on a wavelet filtering algorithm and a noise threshold to obtain a filtered signal;

[0027] S27, removing the bands in which the peak signal intensity of the main wave signal or the echo signal in the filtered signal is less than the low signal threshold, so as to obtain non-low intensity signals.

[0028] Preferably, in step S3, the following steps are specifically included:

[0029] S31, extracting the echo waveform of each band of the non-low-intensity signal;

[0030] S32, obtaining the maximum quantized intensity value of the laser radar echo signal receiver;

[0031] S33, marking the echo waveform with the maximum quantized intensity value of the signal intensity as a saturated waveform;

[0032] S34. Classify the echo waveform after removing the saturated waveform based on the random forest model to obtain a normal waveform with a clear peak and a U-shaped waveform with distortion between the rising edge and the falling edge.

[0033] Preferably, in step S34, the following steps are specifically included:

[0034] S341, normalize each band of the echo waveform after removing the saturated waveform to obtain the waveform to be classified; the normalization calculation formula is:

[0035]

[0036] In the above formula, I normal represents the signal strength of the normalized echo waveform, I represents the signal strength of the echo waveform, max(I) and min(I) represent the maximum and minimum signal strengths in the time domain waveform of the echo waveform, respectively;

[0037] S342, setting sample data of a normal waveform with an obvious peak value and a U-shaped waveform with distortion between the rising edge and the falling edge;

[0038] S343, setting a random forest model and training it through sample data;

[0039] S344. Classify the waveform to be classified by using the trained random forest model to obtain a normal waveform and a U-shaped waveform.

[0040] Preferably, in step S4, the following steps are specifically included:

[0041] S41. Calculate the waveform arrival time of the normal waveform by the peak value method; the calculation formula of the waveform arrival time of the normal waveform is:

[0042]

[0043] In the above formula, t p Indicates the peak moment of the laser signal waveform, t i represents the sampling time corresponding to the waveform sampling point i, f(t i ) represents the quantized value of the signal strength at waveform sampling point i;

[0044] S42, calculating the waveform arrival time of the saturation waveform by a waveform fitting method;

[0045] S43, calculating the waveform arrival time of the U-shaped waveform by the centroid method; the calculation formula of the waveform arrival time of the U-shaped waveform is:

[0046]

[0047] In the above formula, t c represents the time corresponding to the centroid position of the U-shaped waveform, f(t i ) represents the quantized value of the signal strength at the waveform sampling point i corresponding to the sampling time, t i represents the sampling time corresponding to the waveform sampling point i, and k represents the number of sampling times of the U-shaped waveform.

[0048] Preferably, in step S42, the following steps are specifically included:

[0049] S421. Perform waveform fitting on the saturated waveform by using a skewed Gaussian function. The calculation formula for waveform fitting is as follows:

[0050]

[0051] In the above formula, f(t) represents the waveform signal of the saturated waveform, A represents the amplitude, τ represents the center position of the saturated waveform, and ω represents the half-height width of the saturated waveform;

[0052] S422. Calculate the waveform arrival time of the saturated waveform after waveform fitting by using the peak value method.

[0053] Preferably, in step S5, the following steps are specifically included:

[0054] S51, obtaining the signal emission time of the main wave signal of each band of the non-low-intensity signal and the waveform arrival time of the corresponding echo signal;

[0055] S52, calculating the first target distance between the target and the signal receiving position according to the signal transmission time, waveform arrival time and light speed of each band of the non-low-intensity signal; the calculation formula of the first target distance is:

[0056]

[0057] In the above formula, The λth non-low-intensity signal j The first target distance between the target calculated by each band and the signal receiving position, C represents the speed of light, and Respectively represent the j-th band λ of the non-low-intensity signal j The signal emission time and waveform arrival time;

[0058] S53, respectively calculating the mode of the first target distance corresponding to each band of the normal waveform, the saturated waveform and the U-shaped waveform in the non-low-intensity signal; the calculation formula of the mode is:

[0059]

[0060] In the above formula, D i represents the mode of the first target distance of the i-th type of waveform, represents the first target of the i-th type of waveform in the j-th band, n i represents the number of bands classified as the i-th waveform;

[0061] S54, calculating the target distance according to the mode of the first target distance corresponding to each band of the normal waveform, the saturated waveform and the U-shaped waveform in the non-low-intensity signal; the calculation formula of the target distance is:

[0062]

[0063] In the above formula, D represents the target distance, and N represents the number of bands involved in waveform classification.

[0064] The technical solution also provides a system for implementing a ranging method based on a laser radar saturation signal, the system comprising: a processor and a memory, the memory being used to store a computer program, and the computer program implementing any one of the ranging methods when executed by the processor.

[0065] Compared with the prior art, the present invention provides a ranging method and system based on laser radar saturation signal, which has the following beneficial effects:

[0066] 1. The present invention eliminates noise signals and low-intensity signals from waveform data to obtain non-low-intensity signals, thereby ensuring the accuracy of waveform data, and then classifies the echo signals in the non-low-intensity signals, so as to use different time identification algorithms to calculate the waveform arrival time with higher accuracy, and then calculates the target distance according to the emission time of the main wave signal of each band, the waveform arrival time of the echo signal and the speed of light.

[0067] 2. The present invention obtains the background noise signal according to the ending time of the falling edge of the main wave signal and the starting time of the rising edge of the echo signal in the corresponding band, and calculates the noise threshold and the low signal threshold by calculating the average value and standard deviation of the background noise signal to complete the subsequent signal filtering and remove the low-intensity signal.

[0068] 3. The present invention identifies the saturated signal in the echo waveform based on the maximum quantized intensity value of the laser radar echo signal receiver, and further classifies the echo waveform after removing the saturated waveform according to the random forest model, so as to facilitate the subsequent calculation of the waveform arrival time according to the corresponding type of waveform.

[0069] 4. The present invention calculates the first target distance of each type of waveform in each band, and further calculates the mode of the first target distance of each type of waveform, and finally calculates the target distance in a weighted manner, so that the obtained target distance has higher accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] The drawings described herein are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0071] Figure 1 It is a flow chart of the distance measurement method based on the laser radar saturation signal of the present invention;

[0072] Figure 2 A first line graph comparing the distance measurement accuracy of the present invention and the traditional method;

[0073] Figure 3 A second line graph comparing the distance measurement accuracy of the present invention and the traditional method;

[0074] Figure 4 A third line graph comparing the distance measurement accuracy of the present invention and the traditional method;

[0075] Figure 5 is a schematic diagram of a saturation waveform of the present invention;

[0076] Figure 6 is a schematic diagram of a normal waveform of the present invention;

[0077] Figure 7 It is a schematic diagram of the U-shaped waveform of the present invention. DETAILED DESCRIPTION

[0078] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific implementation methods, so that the implementation process of how the present application uses technical means to solve technical problems and achieve technical effects can be fully understood and implemented accordingly.

[0079] Those skilled in the art can understand that all or part of the steps in the following embodiments can be completed by instructing the relevant hardware through a program, so the present application can be in the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present application can be in the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program codes.

[0080] In order to solve the problem that the current TOF ranging method cannot adapt to the time error caused by various waveform distortions when measuring the distance of highly reflective targets, thereby reducing the accuracy of distance measurement, the present invention provides a ranging method based on a saturated laser radar signal to reduce the measurement error of the flight time caused by distorted signals such as saturated echoes, thereby improving the ranging accuracy. Figure 1 As shown, the ranging method includes the following steps:

[0081] S1. Collecting waveform data containing multiple bands based on a hyperspectral laser radar, wherein each band of the waveform data contains a main wave signal emitted by the laser and an echo signal reflected by a target;

[0082] S2, obtaining the noise signal between the end time of the main wave signal and the start time of the echo signal in each band to calculate its average value and the sum of several times the standard deviation to obtain the noise threshold and the low signal threshold, and filtering the waveform data and removing the low intensity signal to obtain the non-low intensity signal, and the target distance can be calculated by calculating the time difference between the main wave signal and the echo signal. In step S2, the following steps are specifically included:

[0083] S21, obtaining the ending time of the falling edge of the main wave signal in each band and the starting time of the rising edge of the echo signal in the corresponding band;

[0084] S22, acquiring waveform data between the end time of the falling edge of the main wave signal and the start time of the rising edge of the echo signal in the corresponding band in the waveform data, and marking it as background noise;

[0085] S23, calculating the average value of the noise signal of the background noise; the calculation formula of the noise signal average value is:

[0086]

[0087] In the above formula, μ noise Indicates the average value of the noise signal, wavenoise i represents the noise signal strength at the sampling time corresponding to the i-th sampling point, and n represents a total of n sampling points;

[0088] S24, calculating the standard deviation of background noise; the calculation formula of the standard deviation of background noise is:

[0089]

[0090] In the above formula, σ noise represents the standard deviation of background noise;

[0091] S25, obtaining a noise threshold and a low signal threshold according to the average value of the background noise and the sum of several times the standard deviation;

[0092] The noise threshold is calculated as:

[0093] Noise threshold =μ noise +w Noise σ noise

[0094] The calculation formula for the low signal threshold is:

[0095] LowSignal threshold =μ noise +w LowSignal σ noise

[0096] In the above formula, Noise threshold represents the noise threshold, w Noise Represents the noise factor, LowSignal threshold Indicates the low signal threshold, w LowSignal Indicates a low signal coefficient. Generally, the noise coefficient w Noise and low signal coefficient w LowSignal Setting it to 3 can achieve better subsequent filtering and removal of low-intensity signals;

[0097] S26, filtering out noise in the waveform data based on a wavelet filtering algorithm and a noise threshold to obtain a filtered signal;

[0098] S27, removing the band in which the peak signal intensity of the main wave signal or the echo signal in the filtered signal is less than the low signal threshold, that is, removing the low intensity signal to obtain a non-low intensity signal.

[0099] S3. Classify each echo waveform of the non-low-intensity signal into a saturated waveform with obvious truncation, a normal waveform with obvious peaks, and a U-shaped waveform with distortion between the rising edge and the falling edge. In actual operation, in order to ensure the accuracy of the echo waveform, the echo waveform signal after filtering can be filtered with a value greater than the noise threshold Noise. threshold The first data point is taken as the starting point, and the waveform is extracted with a fixed time length of 6ns, that is, 30 waveform data sampling points are taken as a complete waveform. For different types of waveforms, different time identification algorithms need to be used to calculate the waveform arrival time. In step S3, the following steps are specifically included:

[0100] S31, extracting the echo waveform of each echo of the non-low-intensity signal;

[0101] S32, obtaining the maximum quantized intensity value of the laser radar echo signal receiver, such as Figure 1 255 as shown in;

[0102] S33, marking the echo waveform with the maximum quantized intensity value of the signal intensity as a saturated waveform;

[0103] S34. Based on the random forest model, the echo waveform after removing the saturated waveform is classified to obtain a normal waveform with a clear peak and a U-shaped waveform with distortion between the rising edge and the falling edge. For the normal waveform and the U-shaped waveform, the schematic diagrams of different waveforms are as follows: Figure 5-Figure 7 As shown, the intensity of the waveform signal varies with the distance and the target material, so we need to first normalize the waveform of each band to avoid the influence of the waveform intensity on the classification. In step S34, the following steps are specifically included:

[0104] S341, normalize each band of the echo waveform after removing the saturated waveform to obtain the waveform to be classified; the normalization calculation formula is:

[0105]

[0106] In the above formula, I normal represents the signal strength of the normalized echo waveform, I represents the signal strength of the echo waveform, max(I) and min(I) represent the maximum and minimum signal strengths in the time domain waveform of the echo waveform, respectively;

[0107] S342, setting sample data of a normal waveform with an obvious peak value and a U-shaped waveform with distortion between the rising edge and the falling edge;

[0108] S343, setting a random forest model and training it through sample data;

[0109] S344. Classify the waveform to be classified by using the trained random forest model to obtain a normal waveform and a U-shaped waveform.

[0110] S4, respectively using the peak method, waveform fitting method and centroid method to calculate the waveform arrival time of the normal waveform, saturated waveform and U-shaped waveform. For different echo waveforms, different time discrimination methods need to be used to calculate the waveform arrival time with higher accuracy. In step S4, the following steps are specifically included:

[0111] S41. Calculate the waveform arrival time of the normal waveform by the peak value method; the calculation formula of the waveform arrival time of the normal waveform is:

[0112]

[0113] In the above formula, t p Indicates the peak moment of the laser signal waveform, t i represents the sampling time corresponding to the waveform sampling point i, f(t i ) represents the quantized value of the signal strength at waveform sampling point i;

[0114] S42, calculating the waveform arrival time of the saturated waveform by a waveform fitting method; in order to better achieve the waveform fitting of the saturated waveform in the present invention and to more accurately calculate the waveform arrival time, specifically, in step S42, the following steps are included:

[0115] S421. Perform waveform fitting on the saturated waveform by using a skewed Gaussian function. The calculation formula for waveform fitting is as follows:

[0116]

[0117] In the above formula, f(t) represents the waveform signal of the saturated waveform, A represents the amplitude, τ represents the center position of the saturated waveform, and ω represents the half-height width of the saturated waveform;

[0118] S422. Calculate the waveform arrival time of the saturated waveform after waveform fitting by using the peak value method.

[0119] S43. For the U-shaped echo waveform, since the peak characteristics are unclear, the waveform arrival time of the U-shaped waveform is calculated by the centroid method; the calculation formula of the waveform arrival time of the U-shaped waveform is:

[0120]

[0121] In the above formula, t c represents the time corresponding to the centroid position of the U-shaped waveform, f(t i ) represents the quantized value of the signal strength at the waveform sampling point i corresponding to the sampling time, t irepresents the sampling time corresponding to the waveform sampling point i, and k represents the number of sampling times of the U-shaped waveform.

[0122] S5. Calculate the target distance between the target and the signal receiving position according to the waveform arrival time of the echo waveform, the emission time of the main wave signal and the speed of light. Since the target distances calculated by different echo waveforms in different bands may be different, they need to be processed to calculate the target distance with higher accuracy. In step S5, the following steps are specifically included:

[0123] S51, obtaining the signal emission time of the main wave signal of each band of the non-low-intensity signal and the waveform arrival time of the corresponding echo signal;

[0124] S52, calculating the first target distance between the target and the signal receiving position according to the signal transmission time, waveform arrival time and light speed of each band of the non-low-intensity signal; the calculation formula of the first target distance is:

[0125]

[0126] In the above formula, The λth non-low-intensity signal j The first target distance between the target calculated by each band and the signal receiving position, C represents the speed of light, and Respectively represent the j-th band λ of the non-low-intensity signal j The signal emission time and waveform arrival time;

[0127] S53, respectively calculating the mode of the first target distance corresponding to each band of the normal waveform, the saturated waveform and the U-shaped waveform in the non-low-intensity signal; the calculation formula of the mode is:

[0128]

[0129] In the above formula, D i represents the mode of the first target distance of the i-th type of waveform, represents the first target of the i-th type of waveform in the j-th band, n i represents the number of bands classified as the i-th waveform;

[0130] S54, calculating the target distance according to the mode of the first target distance corresponding to each band of the normal waveform, the saturated waveform and the U-shaped waveform in the non-low-intensity signal; the calculation formula of the target distance is:

[0131]

[0132] In the above formula, D represents the target distance, and N represents the number of bands involved in waveform classification.

[0133] In order to verify the actual effect of the present invention, "Watch out for pedestrians" traffic signs were collected at distances of 3m-12m with a step length of 0.5m. The collection was repeated three times at each measuring distance. The distance between the target and the hyperspectral lidar was measured with a tape measure as the reference distance. The multi-band ranging method proposed in this patent was used for ranging (denoted as RMWC). To illustrate the effect of the invention, the background technology used was to apply the peak method, centroid method and waveform fitting method (denoted as PV-905, CM-905, WF-905) to process the echo data under the 905nm band, and compared the results of the single-band ranging method obtained at 905nm. The standard deviation, average error and maximum deviation were used to evaluate the ranging effects of different methods. The results showed that for traffic signs, the ranging results of RMWC were relatively stable and the error was kept at a low level. Overall, RMWC had better ranging accuracy and more stable ranging effects than the traditional single-band ranging algorithm. The line graph of its statistical results is shown in the figure. Figure 2 , Figure 3 and Figure 4 shown.

[0134] Corresponding to the ranging method based on the laser radar saturation signal provided in the above embodiment, the present embodiment also provides a system for implementing the ranging method based on the laser radar saturation signal. Since the ranging system based on the laser radar saturation signal provided in the present embodiment corresponds to the ranging method based on the laser radar saturation signal provided in the above embodiment, the implementation method of the aforementioned ranging method based on the laser radar saturation signal is also applicable to the ranging system based on the laser radar saturation signal provided in the present embodiment, and will not be described in detail in the present embodiment.

[0135] The system includes a processor and a memory, wherein the memory is used to store a computer program, and when the computer program is executed by the processor, a ranging method based on a laser radar saturation signal is implemented.

[0136] The above implementation methods have been described in detail. Specific examples are used herein to illustrate the principles and implementation methods of the present invention. The description of the above embodiments is only used to help understand the method of the present invention and its core idea. At the same time, for those skilled in the art, according to the idea of ​​the present invention, there will be changes in the specific implementation methods and application scope. In summary, the content of this specification should not be understood as a limitation on the present invention.

Claims

1. A ranging method based on laser radar saturation signal, characterized in that: The ranging method comprises the following steps: S1. Collecting full waveform data containing multiple bands based on a hyperspectral laser radar, wherein each band of the waveform data contains a main wave signal emitted by the laser and an echo signal reflected by a target; S2, obtaining the noise signal between the end time of the main wave signal and the start time of the echo signal in each band to calculate its average value and the sum of several times of the standard deviation as the noise threshold and the low signal threshold, and filtering the multi-band data and removing the low-intensity band to obtain a non-low-intensity signal; S3, classifying each echo waveform of the non-low-intensity signal into a saturated waveform with obvious waveform truncation, a normal waveform with obvious peaks, and a U-shaped waveform with distortion between the rising edge and the falling edge; S4, respectively calculating the waveform arrival time of the normal waveform, the saturated waveform and the U-shaped waveform by using the peak method, the waveform fitting method and the centroid method; S5. Calculate the target distance between the target and the signal receiving position according to the waveform arrival time of the echo waveform, the emission time of the main wave signal and the speed of light.

2. The distance measurement method according to claim 1, characterized in that: In step S2, the following steps are specifically included: S21, obtaining the ending time of the falling edge of the main wave signal in each band and the starting time of the rising edge of the echo signal in the corresponding band; S22, acquiring waveform data between the end time of the falling edge of the main wave signal and the start time of the rising edge of the echo signal in the corresponding band in the waveform data, and marking it as background noise; S23, calculating the average value of the noise signal of the background noise; the calculation formula of the noise signal average value is: In the above formula, μ noise Indicates the average value of the noise signal, wavenoise i represents the noise signal strength at the sampling time corresponding to the i-th sampling point, and n represents a total of n sampling points; S24, calculating the standard deviation of background noise; The standard deviation of background noise is calculated as: In the above formula, σ noise represents the standard deviation of background noise; S25, obtaining a noise threshold and a low signal threshold according to the average value of the background noise and the sum of several times the standard deviation; The noise threshold is calculated as: Noise threshold =μ noise +w Noise s noise The calculation formula for the low signal threshold is: LowSignal threshold =μ noise +w LowSignal σ noise In the above formula, Noise threshold represents the noise threshold, w Noise Represents the noise factor, LowSignal threshold Indicates the low signal threshold, w LowSignal Indicates low signal coefficient; S26, filtering out noise in the waveform data based on a wavelet filtering algorithm and a noise threshold to obtain a filtered signal; S27, removing the bands in which the peak signal intensity of the main wave signal or the echo signal in the filtered signal is less than the low signal threshold, so as to obtain non-low intensity signals.

3. The distance measurement method according to claim 1, characterized in that: In step S3, the following steps are specifically included: S31, extracting the echo waveform of each band of the non-low-intensity signal; S32, obtaining the maximum quantized intensity value of the laser radar echo signal receiver; S33, marking the echo waveform with the maximum quantized intensity value of the signal intensity as a saturated waveform; S34. Classify the echo waveform after removing the saturated waveform based on the random forest model to obtain a normal waveform with a clear peak and a U-shaped waveform with distortion between the rising edge and the falling edge.

4. The distance measurement method according to claim 3, characterized in that: In step S34, the following steps are specifically included: S341, normalize each band of the echo waveform after removing the saturated waveform to obtain the waveform to be classified; the normalization calculation formula is: In the above formula, I normal represents the signal strength of the normalized echo waveform, I represents the signal strength of the echo waveform, max(I) and min(I) represent the maximum and minimum signal strengths in the time domain waveform of the echo waveform, respectively; S342, setting sample data of a normal waveform with an obvious peak value and a U-shaped waveform with distortion between the rising edge and the falling edge; S343, setting a random forest model and training it through sample data; S344. Classify the waveform to be classified by using the trained random forest model to obtain a normal waveform and a U-shaped waveform.

5. The distance measurement method according to claim 1, characterized in that: In step S4, the following steps are specifically included: S41. Calculate the waveform arrival time of the normal waveform by the peak value method; the calculation formula of the waveform arrival time of the normal waveform is: In the above formula, t p Indicates the peak moment of the laser signal waveform, t i represents the sampling time corresponding to the waveform sampling point i, f(t i ) represents the quantized value of the signal strength at waveform sampling point i; S42, calculating the waveform arrival time of the saturation waveform by a waveform fitting method; S43, calculating the waveform arrival time of the U-shaped waveform by the centroid method; the calculation formula of the waveform arrival time of the U-shaped waveform is: In the above formula, t c represents the time corresponding to the centroid position of the U-shaped waveform, f(t i ) represents the quantized value of the signal strength at the waveform sampling point i corresponding to the sampling time, t i represents the sampling time corresponding to the waveform sampling point i, and k represents the number of sampling times of the U-shaped waveform.

6. The distance measurement method according to claim 5, characterized in that: In step S42, the following steps are specifically included: S421. Perform waveform fitting on the saturated waveform by using a skewed Gaussian function. The calculation formula for waveform fitting is as follows: In the above formula, f(t) represents the waveform signal of the saturated waveform, A represents the amplitude, τ represents the center position of the saturated waveform, and ω represents the half-height width of the saturated waveform; S422. Calculate the waveform arrival time of the saturated waveform after waveform fitting by using the peak value method.

7. The distance measurement method according to claim 1, characterized in that: In step S5, the following steps are specifically included: S51, obtaining the signal emission time of the main wave signal of each band of the non-low-intensity signal and the waveform arrival time of the corresponding echo signal; S52, calculating the first target distance between the target and the signal receiving position according to the signal transmission time, waveform arrival time and light speed of each band of the non-low-intensity signal; the calculation formula of the first target distance is: In the above formula, The λth non-low-intensity signal j The first target distance between the target calculated by each band and the signal receiving position, C represents the speed of light, and Respectively represent the j-th band λ of the non-low-intensity signal j The signal emission time and waveform arrival time; S53, respectively calculating the mode of the first target distance corresponding to each band of the normal waveform, the saturated waveform and the U-shaped waveform in the non-low-intensity signal; the calculation formula of the mode is: In the above formula, D i represents the mode of the first target distance of the i-th type of waveform, represents the first target of the i-th type of waveform in the j-th band, n i represents the number of bands classified as the i-th waveform; S54, calculating the target distance according to the mode of the first target distance corresponding to each band of the normal waveform, the saturated waveform and the U-shaped waveform in the non-low-intensity signal; the calculation formula of the target distance is: In the above formula, D represents the target distance, and N represents the number of bands involved in waveform classification.

8. A system for implementing the distance measurement method according to any one of claims 1 to 7, characterized in that: The method comprises a processor and a memory, wherein the memory is used to store a computer program, and when the computer program is executed by the processor, the ranging method according to any one of claims 1 to 7 is implemented.

Images