A time identification method for multi-transmit and multi-receive panoramic laser radar

Abstract

The present invention discloses a time identification method applicable to a multi-transmitter and multi-receiver panoramic laser radar. The method determines whether the original echo signal received in each quadrant contains a target echo based on the original echo signal received. When the original echo signal contains a target echo, the original echo signal is fully waveform sampled to obtain a digital signal data set. Subsequently, the digital signal data set is preprocessed. A feature extraction method is used on the preprocessed signal to reduce the complexity of the algorithm. A maximum value detection method is used to determine the number of peaks in the feature signal and the range of the target echo return time. The target echo return time is determined using a peak detection method. The present invention can achieve target presence discrimination, backscatter recognition, target echo peak extraction, and determination of the target echo return time without complex signal processing. The method has higher time identification accuracy, can avoid misjudgments or false alarms, can be used in various environments such as smoke and dust, and has good environmental adaptability and robustness.

Description

Technical Field

[0001] The present invention belongs to the field of laser detection, and in particular relates to a time identification method applicable to a multi-transmitting and multi-receiving circumferential laser radar. Background Art

[0002] Laser detection has the advantages of strong directionality, strong resistance to electromagnetic interference, and high ranging accuracy, making it widely used in various ammunition detection systems. Laser proximity fuzes measure the distance between the projectile and the target by measuring the time interval between the laser pulse emitted and the received pulse signal. The laser fuze's range setting can be easily adjusted by adjusting the set target time value. The accuracy of laser ranging is closely related to many factors, including timing accuracy and time identification accuracy. Excessive errors in the return echo time obtained by time identification can seriously affect the fuze's ranging function. Therefore, research on how to improve time identification accuracy is of great significance.

[0003] The accuracy of time identification is significantly affected by the echo waveform. During the detection process, laser fuzes are affected by factors such as the properties of the emitted laser, target distance, target optical properties, and the transmission characteristics of the environment. Consequently, the amplitude and shape of the echo signal can vary significantly. For example, in environments such as fog, smoke, and dust, the echo signal is superimposed with backscatter, which can affect the system's recognition of the target echo. When the target's reflective properties are too strong, the echo signal saturates, resulting in saturation drift errors. Therefore, accurately identifying the echo waveform and extracting the target echo are crucial for improving time identification accuracy.

[0004] The constant ratio timing method ensures accurate time identification even when the echo power fluctuates significantly. However, when the echo signal is saturated or distorted, traditional time identification methods can exhibit significant errors. The leading-edge threshold method uses a fixed threshold, with the start and end times being the moment the signal reaches the threshold, to achieve ranging. However, if backscatter is superimposed on the target echo signal, the leading-edge threshold method will ultimately misidentify the backscatter echo as the target echo, resulting in significant time identification errors. To improve time identification accuracy, numerous institutions have conducted in-depth research and proposed a variety of new time identification methods. Wang Cheng designed a time discrimination circuit based on the constant ratio delay time discrimination method. By calibrating the relationship between the time difference between the two thresholds of the rising edge of the signal and the time drift error, the time drift error caused by the fixed threshold was eliminated (WANGC.Lidar ranging system based on constant ratio delay time discrimination method[J].Information Technology and Informatization.2015(10):195-196); Chen Ruiqiang proposed a double threshold leading edge time discrimination method, which eliminated the echo time drift error and improved the accuracy of single threshold time discrimination, thereby meeting the requirements of high-precision and high-frequency ranging (CHEN RQ, JIANG YS, PEI Z.High frequency and high accuracy laser ranging system based on double thresholds leading-edge timing discrimination[J].Acta Optica Sinica,2013,33(9):155-162); Wu Yu proposed a time discrimination method using constant threshold and peak dual channels, established a time domain distribution model of the echo waveform, and realized time discrimination that was not affected by signal broadening and attenuation (WU Y, ZHOU MC, ZHAO Q, et al. al.Threshold—peakdual-channel time discrimination method for pulse laser ranging[J].Infraredand Laser Engineering.2019,48(06):318-324). At present, certain achievements have been made in improving the accuracy of time identification. However, in complex detection environments, the laser fuze will receive superimposed backscattered echo signals. The signal processing process of the current time identification method is relatively cumbersome and cannot meet the needs of identifying backscattered echoes and extracting target echoes in a short time to improve the accuracy of time identification. Summary of the Invention

[0005] The present invention proposes a time identification method suitable for a multi-transmitter and multi-receiver panoramic laser radar, which can realize target presence or absence discrimination, backscatter recognition and target echo peak extraction without complex signal processing, thereby improving the time identification accuracy.

[0006] The technical solution to realize the present invention is: a time identification method applicable to a multi-transmitter and multi-receiver panoramic laser radar, the steps of which are as follows:

[0007] Step 1: Use a multi-transmitter, multi-receiver panoramic lidar to emit laser light, and each quadrant receives the original echo signal. Then, use the multi-transmitter, multi-receiver panoramic lidar to perform full waveform sampling on the original echo signal to obtain the digital signal data set corresponding to each quadrant, and then proceed to step 2.

[0008] There are several situations for the original echo signal: 1) only the target echo signal, 2) no target echo signal, 3) the target echo signal containing interference, and 4) only the interference signal.

[0009] Step 2: Determine whether the digital signal dataset of each quadrant of the multi-transmitter and multi-receiver panoramic lidar contains the target echo signal. If the digital signal dataset contains the target echo signal, that is, it is only the target echo signal or the target echo signal containing interference, then go to step 3; otherwise, return to step 1.

[0010] Step 3: Perform saturation echo detection on the digital signal data set. If a saturation echo is detected, calculate the start and end times of the saturation band to obtain the target echo return time, and proceed to step 8. If no saturation echo is detected, proceed to step 4.

[0011] Step 4: Preprocess the digital signal data set to filter out noise, obtain the preprocessed signal, and proceed to step 5:

[0012] First, the digital signal data set is subjected to a limiting filter method to obtain a limited-filtered signal. Then, the limited-filtered signal is subjected to a recursive averaging filter method to obtain a recursive average filtered signal. Finally, the recursive average filtered signal is subjected to a threshold truncation method to obtain a preprocessed signal.

[0013] Step 5: Store the pre-processed signal in the FIFO. Write the pre-processed signal stored in the FIFO into the DDR through the AXI-Full interface and proceed to step 6.

[0014] Step 6: Use feature extraction method on the pre-processed signal to obtain the characteristic signal. Perform maximum value detection method and statistical discrimination method on the characteristic signal to obtain the target echo peak in the characteristic signal, and then go to step 7:

[0015] To obtain the shape characteristics of the preprocessed signal, a feature extraction method is applied to the preprocessed signal to generate a characteristic signal. Based on this characteristic signal, a maximum detection method is used to identify the maximum points within the characteristic signal, resulting in a maximum point set. A statistical discriminant method is then used to further filter these points within the maximum point set. This process results in one or two qualified maximum points. The number of qualified maximum points identified after screening can be used to determine the target echo peak within the characteristic signal.

[0016] Step 7: Determine the target echo return time range in the characteristic signal based on the target echo peak in the characteristic signal. The target echo return time range in the preprocessed signal is determined from the mapping relationship between the characteristic signal and the preprocessed signal. Peak detection is used within the target echo return time range in the preprocessed signal to determine the target echo return time, and the process then proceeds to Step 8.

[0017] Step 8. Return to step 1 for the next scan.

[0018] Compared with the prior art, the present invention has the following significant advantages:

[0019] (1) The present invention can realize the discrimination of the presence or absence of a target, the recognition of backscattering, the extraction of the target echo peak, and the determination of the target echo return time without complex signal processing; it can adapt to various environments such as smoke and dust and has high robustness.

[0020] (2) The present invention uses a feature extraction method to retain the shape characteristics of the preprocessed signal, reducing the complexity of the moment identification algorithm without reducing the accuracy of moment identification; and uses a maximum value detection method and a statistical discrimination method to realize backscatter echo recognition and target echo peak extraction.

[0021] (3) The present invention adopts the continuous saturation discrimination method to realize the recognition of saturated echoes, and determines the return time of the echo by detecting the start and end time of the saturated band, which effectively improves the recognition ability of the laser fuze for saturated echoes and the time identification accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] Figure 1 The present invention is a block diagram of a method for implementing a time identification method for a multi-transmit and multi-receive panoramic laser radar.

[0023] Figure 2 This is a flow chart of target echo signal discrimination of the present invention.

[0024] Figure 3 Schematic diagram of the saturation echo detection method of the present invention.

[0025] Figure 4 This is a diagram of the pretreatment effect of the present invention.

[0026] Figure 5This is the effect diagram of the feature extraction method of the present invention.

[0027] Figure 6 Schematic diagram of the maximum value detection method of the present invention.

[0028] Figure 7 This is a schematic diagram of the statistical discrimination method used in the present invention.

[0029] Figure 8 Schematic diagram of the locking range of the present invention.

[0030] Figure 9 This is a diagram showing the effect of moment identification of a saturated echo signal in an embodiment of the present invention.

[0031] Figure 10 This is a diagram showing the effect of distinguishing the time without superimposing the backscattered echo signal in an embodiment of the present invention.

[0032] Figure 11 This is a diagram showing the effect of time identification for superimposed backscattered echo signals in an embodiment of the present invention. DETAILED DESCRIPTION

[0033] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0034] Combine Figures 1 to 8 A time identification method for a multi-transmitter and multi-receiver panoramic laser radar is provided. The specific implementation steps are as follows:

[0035] Step 1: Use a multi-transmitter, multi-receiver panoramic lidar to emit laser light. Each quadrant receives the original echo signal. Then, use the multi-transmitter, multi-receiver system to perform full waveform sampling on the original echo signal to obtain the digital signal data set corresponding to each quadrant, and then proceed to step 2:

[0036] Combine Figure 1 The multi-transmitter and multi-receiver panoramic lidar periodically emits laser pulses; at the same time, the multi-transmitter and multi-receiver panoramic lidar begins to perform full waveform sampling on the original echo signal according to the set sampling number, and obtains the digital signal data set corresponding to each quadrant, where the sampling number is set according to the sampling frequency and the distance to be detected.

[0037] There are several situations for the original echo signal: 1) only the target echo signal, 2) no target echo signal, 3) the target echo signal containing interference, and 4) only the interference signal.

[0038] Step 2: Determine whether the digital signal dataset of each quadrant of the multi-transmitter and multi-receiver panoramic lidar contains the target echo signal. If the digital signal dataset contains the target echo signal (only the target echo signal or the target echo signal containing interference), proceed to step 3; otherwise, return to step 1.

[0039] Combine Figure 2 Taking the four quadrants as an example, if the four quadrants of the multi-transmitter and multi-receiver panoramic lidar have the same digital signal data set, it means that the digital signal data set only contains interference signals or no target echo signals. Otherwise, it means that the digital signal data set contains target echo signals.

[0040] The processing of digital signal data sets is a non-real-time signal processing method; the non-real-time signal processing method can more comprehensively obtain the characteristics of the digital signal data sets and avoid misjudgment caused by the superposition of backscatter in the original echo signal. Therefore, the non-real-time signal processing method can improve the accuracy of moment identification.

[0041] Step 3: Perform saturation echo detection on the digital signal data set. If a saturation echo is detected, calculate the start and end times of the saturation band to obtain the target echo return time, and proceed to step 8; if no saturation echo is detected, proceed to step 4.

[0042] Combine Figure 3 , the continuous saturation discrimination method is used to realize saturation echo detection on the digital signal data set: if there are multiple consecutive discrete data in the digital signal data set that reach the saturation value, the target echo signal represented by the digital signal data set is determined to be a saturation echo, and the start and end times of the saturation band are continued to be detected, and finally the average of the start and end times is taken as the return time of the target echo.

[0043] Step 4: In order to filter out some noise, the digital signal data set is preprocessed to obtain a preprocessed signal.

[0044] Combine Figure 4 In order to filter out noise with large amplitude variations (the difference between two adjacent digital signal data exceeds 5% of the maximum value of the digital signal data), the digital signal data set is first subjected to the limiting filtering method to obtain the limited filtered signal; in order to filter out noise with small amplitude variations (except for noise with large amplitude variations) and improve the smoothness of the signal, the limited filtered signal is then subjected to the recursive averaging filtering method to obtain the recursive averaging filtered signal; in order to remove non-target and non-backscattering band noise, the recursive averaging filtered signal is finally subjected to the threshold truncation method to obtain the preprocessed signal.

[0045] S4.1. Apply a limiting filtering method to the digital signal data set to obtain a limited filtered signal.

[0046] The calculation method of the limiting filter method is as follows:

[0047]

[0048] Where y i is the i-th digital signal data; w i is the i-th data after limiting filtering; A is the set maximum allowable deviation. The limiting filter is calculated by calculating the deviation between the current digital signal data and the previous digital signal data. If the deviation does not exceed the maximum allowable deviation, the current digital signal data is considered valid. Otherwise, the current digital signal data is considered invalid and replaced by the previous digital signal data.

[0049] S4.2. Apply recursive averaging filtering to the amplitude-limited filtered signal to obtain a recursive averaging filtered signal.

[0050] The calculation method of the recursive average filter method is as follows:

[0051]

[0052] Where w b is the bth data after limiting filtering; z b is the bth recursive average filtered data; z m+n-2 is the data after the m+n-2th recursive average filtering; z m The mth recursive average filter data; m is the total amount of clipped filtered signal data; n is the number of clipped filtered signal values used in each recursive average filter calculation, which is determined based on the actual situation. The recursive average filter is calculated by averaging each n consecutive clipped filtered signal values and using the average value as the new echo data. n is determined based on the actual situation. Currently, n = 4, and the recursive average filter calculation method is as follows:

[0053]

[0054] Where w c is the cth data after limiting filtering; z c is the cth recursive average filtered data; z m is the data after the mth recursive average filtering; m is the total data volume of the signal after the limiting filtering.

[0055] S4.3. In order to remove non-target and non-backscattering band noise, the threshold truncation method is used on the recursive average filtered signal to obtain the preprocessed signal.

[0056] Combine Figure 4 , the calculation method of the threshold interception method is as follows:

[0057]

[0058] Where v g is the g-th preprocessed data, z g is the gth recursively averaged filtered data, Z is the threshold value, which depends on the actual situation, and m is the total amount of recursively averaged filtered data. The threshold truncation method retains the original recursively averaged filtered data for data that exceeds the threshold, and replaces the original recursively averaged filtered data with the threshold value for data that does not exceed the threshold.

[0059] Go to step 5.

[0060] Step 5: Store the pre-processed signal into the FIFO, write the pre-processed signal stored in the FIFO into the DDR through the AXI-Full interface, and go to step 6.

[0061] Step 6: Use the feature extraction method on the preprocessed signal to obtain the characteristic signal; use the maximum value detection method and statistical discrimination method on the characteristic signal to obtain the target echo peak in the characteristic signal.

[0062] Combine Figure 1 、 Figures 5 to 7 In order to obtain the shape characteristics of the preprocessed signal, the feature extraction method is used on the preprocessed signal to obtain the characteristic signal; on the basis of the characteristic signal, the maximum value detection method is used to obtain the maximum value points in the characteristic signal, and a maximum value point set is obtained; the statistical discrimination method is used to further screen the maximum value points in the maximum value point set, and after screening, 1-2 maximum value points that meet the requirements will be obtained; according to the number of maximum value points that meet the requirements after screening, the target echo peak in the characteristic signal can be determined.

[0063] Combine Figure 5 , the calculation method of the feature extraction method is as follows:

[0064]

[0065] Where v j is the jth preprocessed data, r e is the e-th feature signal data, T is the extraction interval (depending on the actual situation), and m is the total amount of preprocessed signal data. The feature extraction method extracts data from the preprocessed signal at equal intervals as the feature signal data, replacing the shape of the preprocessed signal with the shape of the feature signal; this method preserves the shape characteristics of the preprocessed signal while filtering out a large amount of noise.

[0066] Combine Figure 6 After obtaining the characteristic signal, the maximum value point is detected using the maximum value detection method. First, the slope of each characteristic signal data in the characteristic signal is calculated. The method of using the slope to detect the maximum value point is as follows:

[0067]

[0068] Where k s is the slope of the sth characteristic signal data. After the maximum value detection, a maximum value point set can be obtained.

[0069] Combine Figure 7 Since the original echo signal may contain obvious noise (the variation amplitude is lower than and very close to the noise with large variation amplitude mentioned above, which cannot be filtered out by the limiting filtering method), even after recursive averaging filtering, there is still a certain probability that multiple maximum points will be detected due to incomplete filtering and randomness of noise positions. To avoid large errors in this situation, the statistical discrimination method is used to further screen the maximum points in the maximum point set. The screening criterion is that there must be a certain number of characteristic signal data with data values less than the maximum value on both sides of the maximum point, and the number of data to be screened depends on the actual situation. After using the statistical discrimination method, 1-2 maximum points that meet the conditions can be obtained. If there is only one maximum point that meets the conditions, the peak where the maximum point is located is the target echo peak in the characteristic signal. If there are two maximum points that meet the conditions, it means that backscatter is superimposed on the characteristic signal. Since the target echo lags behind the backscatter, the peak where the second maximum point is located is the target echo peak in the characteristic signal.

[0070] Go to step 7.

[0071] Step 7. Determine the target echo return time range in the characteristic signal based on the target echo peak in the characteristic signal; obtain the target echo return time range in the preprocessed signal from the mapping relationship between the characteristic signal and the preprocessed signal; obtain the target echo return time within the target echo return time range of the preprocessed signal using the peak detection method.

[0072] Combine Figure 8 , the position of the target echo peak in the preprocessed signal must be within the range formed by the target echo peak in the characteristic signal and its two adjacent characteristic signal data; according to the mapping relationship between the characteristic signal and the preprocessed signal in formula (5), the range of the target echo peak in the characteristic signal can be mapped to the range of the target echo peak in the preprocessed signal; within the range where the target echo peak in the preprocessed signal is located, the peak detection method is used on the preprocessed signal to obtain the position of the maximum value in the range; combined with the sampling frequency of the full waveform sampling, the time corresponding to this maximum value can be obtained; the time corresponding to this maximum value can be used as the return time of the target echo in the original echo signal.

[0073] Go to step 8.

[0074] Step 8. Return to step 1 for the next scan.

[0075] Example

[0076] The time identification method of the present invention uses a saturated echo signal, an echo signal without superimposed backscattering, and an echo signal with superimposed backscattering as input data. The time identification effect of the present invention is as follows: Figure 9 、 10 , 11. The fixed delay of this example system is 190ns and the sampling frequency is 1GHz. Figure 9 As shown in the figure, the target is placed at 38.5m, the pulse is transmitted at 200ns, and the system detects a saturated echo. The start and end times of the saturation band are 418ns and 478ns respectively. The average of the two is taken as the return time, which is 448ns. After deducting the fixed delay of the system, the actual return time of the target is 258ns, the distance is 38.7m, and the relative error is 0.5%. Figure 10 As shown in the figure, the target is placed at a position of 52.5m, the pulse emission time is 200ns, the detection return time is 541ns, after deducting the fixed delay, the target's actual return time is 351ns, the distance is 52.65m, and the relative error is 0.29%. Figure 11 As shown, with the target placed at 52.5 meters and the pulse emission time at 200 nanoseconds, the algorithm identifies the second peak as the target echo peak and detects the return time at 539 nanoseconds. After subtracting the fixed delay, the target's true return time is 349 nanoseconds, at a distance of 52.35 meters, with a relative error of 0.29%. After processing using the multi-transmitter, multi-receiver, panoramic lidar timing identification method described herein, the target echo peak is extracted, effectively improving timing identification accuracy.

Claims

1. A time identification method applicable to a multi-transmit and multi-receive panoramic laser radar, characterized in that: Here are the steps: Step 1: Use a multi-transmitter, multi-receiver, panoramic laser radar to emit laser light, and each quadrant receives the original echo signal. Then, the multi-transmitter, multi-receiver, panoramic laser radar performs full waveform sampling on the original echo signal to obtain the digital signal data set corresponding to each quadrant, and then proceed to step 2. The original echo signal has the following situations: 1) only the target echo signal, 2) no target echo signal, 3) the target echo signal containing interference, 4) only the interference signal; Step 2: Determine whether the digital signal dataset of each quadrant of the multi-transmitter and multi-receiver panoramic lidar contains a target echo signal. If the digital signal dataset contains a target echo signal, that is, only a target echo signal or a target echo signal containing interference, proceed to step 3; otherwise, return to step 1. Step 3: Perform saturation echo detection on the digital signal data set. If a saturation echo is detected, calculate the start and end times of the saturation band to obtain the target echo return time, and proceed to step 8. If no saturation echo is detected, proceed to step 4. Step 4: Preprocess the digital signal data set to filter out noise, obtain the preprocessed signal, and proceed to step 5: First, the digital signal data set is subjected to a limiting filtering method to obtain a limiting filtered signal; then, the limiting filtered signal is subjected to a recursive averaging filtering method to obtain a recursive averaging filtered signal; finally, the recursive averaging filtered signal is subjected to a threshold interception method to obtain a preprocessed signal; Step 5: Store the pre-processed signal into FIFO; Write the pre-processed signal stored in the FIFO into the DDR through the AXI-Full interface and go to step 6; Step 6: Perform feature extraction on the pre-processed signal to obtain a feature signal; Perform maximum value detection and statistical discrimination on the characteristic signal to obtain the target echo peak in the characteristic signal, and then proceed to step 7: Step 7: Determine the target echo return time range in the characteristic signal based on the target echo peak in the characteristic signal; obtain the target echo return time range in the preprocessed signal based on the mapping relationship between the characteristic signal and the preprocessed signal; obtain the target echo return time within the target echo return time range of the preprocessed signal using a peak detection method, and proceed to step 8; Step 8. Return to step 1 for the next scan.

2. The time identification method applicable to a multi-transmit and multi-receive circumferential laser radar according to claim 1, characterized in that: In step 1, the multi-transmitter and multi-receiver circumferential laser radar periodically emits laser pulses; at the same time, the multi-transmitter and multi-receiver circumferential laser radar begins to perform full waveform sampling of the original echo signal according to the set sampling number, and obtains a digital signal data set corresponding to each quadrant, where the sampling number is set according to the sampling frequency and the distance to be detected.

3. The time identification method applicable to a multi-transmit and multi-receive panoramic laser radar according to claim 1, characterized in that: In step 3, the continuous saturation discrimination method is used to detect saturated echoes on the digital signal data set: if multiple consecutive discrete data in the digital signal data set reach the saturation value, the target echo signal represented by the digital signal data set is determined to be a saturated echo, and the start and end times of the saturation band are continuously detected. Finally, the average of the start and end times is used as the return time of the target echo.

4. The time identification method applicable to a multi-transmit and multi-receive panoramic laser radar according to claim 1, characterized in that: In step 4, the digital signal data set is preprocessed to filter out noise and obtain a preprocessed signal. The steps are as follows: S4.

1. Apply the limiting filtering method to the digital signal data set to obtain the signal after limiting filtering: The calculation method of the limiting filter method is as follows: Where y i is the i-th digital signal data; w i is the i-th data after limiting filtering; A is the set maximum allowable deviation; the calculation method of limiting filtering is to calculate the deviation between the current digital signal data and the previous digital signal data. If the above deviation does not exceed the maximum allowable deviation, the current digital signal data is considered valid; otherwise, the current digital signal data is considered invalid and the previous digital signal data replaces the current digital signal data; S4.

2. Apply recursive averaging filtering to the signal after amplitude limiting filtering to obtain the recursive averaging filtered signal: The calculation method of the recursive average filter method is as follows: Where w b is the bth data after limiting filtering; z b is the bth recursive average filtered data; z m+n-2 is the data after the m+n-2th recursive average filtering; z m is the mth recursive average filtered data, n is the number of clipped filtered signal values used for each recursive average filter calculation, and the calculation method of the recursive average filter is to average each n consecutive clipped filtered signal values and use the average value as the new echo data: Take n = 4, then the recursive average filter calculation method is as follows: Where w c is the cth data after limiting filtering; z c is the cth recursive average filtered data; z m is the mth recursive average filtered data; S4.

3. In order to remove non-target and non-backscattered band noise, the threshold interception method is applied to the recursive average filtered signal to obtain the preprocessed signal: The calculation method of threshold interception is as follows: Where v g is the g-th preprocessed data, z g is the gth recursive average filtered data, Z is the set threshold, and m is the total data volume of the recursive average filtered signal; The threshold truncation method retains the original recursive average filtered data for recursive average filtered data exceeding the threshold, and replaces the original recursive average filtered data with the threshold for recursive average filtered data not exceeding the threshold.

5. The time identification method applicable to a multi-transmit and multi-receive panoramic laser radar according to claim 1, characterized in that: In step 6, a feature extraction method is applied to the preprocessed signal to obtain a feature signal; The characteristic signal is subjected to the maximum value detection method and the statistical discrimination method to obtain the target echo peak in the characteristic signal, as follows: In order to obtain the shape characteristics of the preprocessed signal, the feature extraction method is used on the preprocessed signal to obtain the characteristic signal; on the basis of the characteristic signal, the maximum value detection method is used to obtain the maximum value points in the characteristic signal, and a maximum value point set is obtained; the statistical discrimination method is used to further screen the maximum value points in the maximum value point set, and after screening, 1-2 maximum value points that meet the requirements will be obtained; according to the number of maximum value points that meet the requirements after screening, the target echo peak in the characteristic signal can be determined.

6. The time identification method applicable to a multi-transmit and multi-receive panoramic laser radar according to claim 5, characterized in that: The feature extraction method is calculated as follows: Where v j is the jth preprocessed data, r e is the e-th characteristic signal data, T is the extraction interval, and m is the total data volume of the preprocessed signal; the feature extraction method extracts the data of the preprocessed signal as the data of the characteristic signal in an equal interval, and replaces the shape of the preprocessed signal with the shape of the characteristic signal; After obtaining the characteristic signal, the maximum value point is detected using the maximum value detection method. First, the slope of each characteristic signal data in the characteristic signal is calculated. The method of using the slope to detect the maximum value point is as follows: Where k s is the slope at the sth characteristic signal data, and a maximum point set can be obtained after maximum value detection; The statistical discrimination method is used to further screen the maximum points in the maximum point set; the screening criterion is that there must be a certain number of characteristic signal data with data values less than the maximum value on both sides of the maximum point. After using the statistical discrimination method, 1-2 maximum points that meet the conditions are obtained. If there is only one maximum point that meets the conditions, the peak where the maximum point is located is the target echo peak in the characteristic signal; if there are two maximum points that meet the conditions, it means that backscattering is superimposed on the characteristic signal. Since the target echo lags behind the backscattering, the peak where the second maximum point is located is the target echo peak in the characteristic signal.

7. The time identification method applicable to a multi-transmit and multi-receive circumferential laser radar according to claim 1, characterized in that: In step 7, the position of the target echo peak in the preprocessed signal must be within the range formed by the target echo peak in the characteristic signal and its two adjacent characteristic signal data; according to the mapping relationship between the characteristic signal and the preprocessed signal, the range of the target echo peak in the characteristic signal can be mapped to the range where the target echo peak in the preprocessed signal is located; Within the range where the target echo peak in the preprocessed signal is located, the peak detection method is used to obtain the maximum value within the range; the time corresponding to this maximum value can be used as the return time of the target echo in the original echo signal.

Images