A data processing method and device for laser radar ranging

Abstract

The invention discloses a data processing method applied to laser radar ranging, comprising: using an IIR filter to filter data with a bit width of W bits from a TDC; counting the first M bits of the W-bit data after filtering to obtain data T1 corresponding to a peak value, and counting BATCH data in total, wherein BATCH is a preset value; selecting data with the first M bits being equal to T1 in the data output by the TDC, counting the middle M bits of these data, obtaining data T2 corresponding to the peak value, and counting BATCH data in total; selecting data with the first 2M bits being equal to [T1, T2] or [T1, T2]-1 in the data output by the TDC, counting the last N bits of these data, obtaining data T3 corresponding to the peak value, and counting 3*BATCH data in total; obtaining a statistical result of [T1, T2, T3]; wherein W is the bit width of the TDC, M and N are preset values, M and N need to satisfy W=M+M+N, and BATCH is jointly determined according to the output frame rate of on-chip statistics and the frequency of TDC input data.

Description

Technical Field

[0001] The present invention belongs to the technical field of laser radar ranging, and more specifically, relates to a data processing method and device applied to laser radar ranging. Background Art

[0002] In a Flash LiDAR system based on Direct Time-Of-Flight (DTOF) technology, there is usually a large amount of distance information generated by a Time-to-Digital Converter (TDC) waiting to be processed. Usually, the data is sent to an off-chip for processing, and the output bandwidth becomes a bottleneck, limiting the measurement speed of the LiDAR and causing a large transmission power consumption. On-chip processing of a large amount of light flight time data can solve the above problems to a certain extent. Therefore, how to develop an efficient data processing solution suitable for on-chip for LiDAR ranging has become an urgent problem to be solved. Summary of the invention

[0003] In view of the defects and improvement needs of the prior art, the present invention provides a data processing solution for laser radar ranging, which can be suitable for efficient on-chip data processing.

[0004] In order to achieve the above object, the present invention provides a data processing method applied to laser radar ranging, comprising:

[0005] (1) Using an IIR filter to filter the W-bit data from the TDC;

[0006] (2) Count the first M bits of the W bits of data after filtering to obtain data T1 corresponding to the peak value, and count BATCH data in total, where BATCH is a preset value;

[0007] (3) In the data output by TDC, select the data whose first M bits are equal to T1, count the middle M bits of these data, and obtain the data T2 corresponding to the peak value, and count BATCH data in total;

[0008] (4) In the data output by TDC, select the data whose first 2M bits are equal to [T1, T2] or [T1, T2]-1, count the last N bits of these data, and obtain the data T3 corresponding to the peak value. A total of 3*BATCH data are counted;

[0009] (5) The statistical result is [T1, T2, T3];

[0010] Where W is the bit width of TDC, M and N are preset values, M and N must satisfy W=M+M+N, and BATCH is determined by the output frame rate of on-chip statistics and the frequency of TDC input data.

[0011] According to another aspect of the present invention, there is also provided a data processing device for laser radar ranging, a filtering processing module, a first statistical module, a second statistical module, a third statistical module and a result output module, wherein:

[0012] The filtering processing module is used to filter the data with a bit width of W bits from the TDC using an IIR filter;

[0013] The first statistical module is used to count the first M bits of the W bits of data after filtering to obtain data T1 corresponding to the peak value, and a total of BATCH data are counted, where BATCH is a preset value;

[0014] The second statistical module is used to select the first M bits of data equal to T1 from the data output by TDC, count the middle M bits of these data, obtain the data T2 corresponding to the peak value, and count BATCH data in total;

[0015] The third statistical module is used to select the first 2M bits of data that are equal to [T1, T2] or [T1, T2]-1 from the data output by the TDC, count the last N bits of these data, obtain the data T3 corresponding to the peak value, and count 3*BATCH data in total;

[0016] The result output module is used to obtain the statistical result [T1, T2, T3];

[0017] Where W is the bit width of TDC, M and N are preset values, M and N must satisfy W=M+M+N, and BATCH is determined by the output frame rate of on-chip statistics and the frequency of TDC input data.

[0018] In general, the above technical solution conceived by the present invention has the following beneficial effects compared with the prior art:

[0019] The present invention provides an efficient data processing solution suitable for on-chip laser radar ranging. The solution first uses a filtering algorithm to eliminate the noise from laser pulse width, SPAD jitter, and TDC circuit signal jitter, and then obtains an accurate TOF value by step-by-step peak and histogram statistics. The solution efficiently implements on-chip statistics of light flight time data from TDC in a Flash laser radar system based on D-ToF technology, greatly reducing the system hardware cost and improving the accuracy and real-time performance of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a flow chart of an efficient data processing solution suitable for on-chip for laser radar ranging provided by the present invention;

[0021] Figure 2 It is a specific implementation example diagram of the efficient data processing solution suitable for on-chip for laser radar ranging provided by the present invention;

[0022] Figure 3 It is a statistical comparison diagram of ToF results before and after filtering of the efficient data processing solution suitable for on-chip for laser radar ranging provided by the present invention; wherein:

[0023] Figure 3 (a) is a statistical histogram of ToF data containing random noise with a mean parameter of 6000 and a standard deviation parameter of 200;

[0024] Figure 3 (b) is the statistical histogram of ToF data after filtering by IIR filter;

[0025] Figure 3 (c) is a statistical histogram of ten results after the ToF data containing random noise with a mean parameter of 6000 and a standard deviation parameter of 200 is processed in steps (2) to (4);

[0026] Figure 3 (d) is a statistical histogram of ten results of ToF data containing random noise with a mean parameter of 6000 and a standard deviation parameter of 200 after being processed by (1) to (4). DETAILED DESCRIPTION

[0027] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0028] like Figure 1 As shown, the present invention provides a data processing method for laser radar ranging, which is suitable for efficient data processing on a chip. The method includes the following steps:

[0029] (1) Using an IIR filter to filter the W-bit data from the TDC;

[0030] The filtering process is used to eliminate the noise from the laser pulse width, SPAD jitter, and TDC circuit signal jitter;

[0031] Specifically, W bits of light flight time data (14 in the embodiment of the present invention) continuously measured by the TDC are read through the handshake protocol, and the data with a bit width of W bits from the TDC are filtered using an IIR (Infinite Impulse Response) digital filter;

[0032] Specifically, some noise from laser pulse width, single photon avalanche diode (SPAD) jitter, and TDC circuit signal jitter is eliminated; the time domain expression of the IIR filter used is: y[k]=0.875·y[k-1]+0.125·x[k].

[0033] (2) Count the first M bits of the W bits of data after filtering to obtain data T1 corresponding to the peak value, and count BATCH data in total, where BATCH is a preset value;

[0034] From the perspective of hardware reuse of the peak statistics module, setting M to 5 can reuse the circuit modules of the statistics module to reduce hardware overhead;

[0035] In this embodiment, the first M bits (5) of the W bits (14) of data after filtering, i.e., the 13th to 9th bits, are counted to obtain data T1 corresponding to the peak value, and a total of BATCH data are counted, wherein the data that can be used for one statistics is equal to the frequency of the TDC input data divided by the output frequency of the statistical module required by the system.

[0036] The data that can be used for one statistics is allocated to two peak detections and one peak statistics. For example, here BATCH data is allocated for each peak detection, and 3*BATCH data is used for peak statistics. That is, BATCH is determined by the following formula: Data used for one statistics = BATCH + BATCH + 3*BATCH = Frequency of TDC input data / Output frequency of statistics module required by the system;

[0037] (3) In the data output by TDC, select the data whose first M bits are equal to T1, count the middle M bits of these data, and obtain the data T2 corresponding to the peak value, and count BATCH data in total;

[0038] In this embodiment, in the data output by TDC, data from the 13th to the 9th bit that are equal to T1 are selected, and data from the 8th to the 4th bit of these data are counted to obtain data T2 corresponding to the peak value, and a total of BATCH data are counted;

[0039] (4) In the data output by TDC, select the data whose first 2M bits are equal to [T1, T2] or [T1, T2]-1, count the last N bits of these data, and obtain the data T3 corresponding to the peak value. A total of 3*BATCH data are counted;

[0040] From the perspective of hardware reuse of the peak statistics module, setting N to 4 can reuse the circuit modules of the statistics module to achieve the effect of reducing hardware overhead;

[0041] In this embodiment, in the data output by TDC, data from the 8th to the 4th bit that is equal to [T1, T2] or [T1, T2]-1 is selected, that is, [T1, T2, 4'b0000] is taken as the center, and the last 4 bits of these data are counted to obtain data T3 corresponding to the peak value, and a total of 3*BATCH data are counted;

[0042] (5) The statistical result is [T1, T2, T3].

[0043] like Figure 2 As shown in the figure, a schematic diagram of the proposed filter filtering and step-by-step peak search is shown. First, the IIR filter filtering can eliminate some noise from the laser pulse width, SPAD jitter, and TDC circuit signal jitter, making the subsequent statistics more accurate. Then, by step-by-step peak detection and fine histogram statistics, the hardware resource overhead is reduced and the real-time performance of the system is improved.

[0044] like Figure 3 As shown, you can see Figure 3 (a) is a statistical histogram of ToF data containing random noise with a mean parameter of 6000 and a standard deviation parameter of 200, that is, a histogram of the original data. Figure 3 (b) for Figure 3 Based on the data in (a), IIR filtering is performed. It can be seen that the data after filtering is more concentrated around 6000, which proves that IIR filter filtering can eliminate some noise from laser pulse width, SPAD jitter, and TDC circuit signal jitter. Figure 3 (c) When BATCH is 20, Figure 3 The data in (a) is filtered ten times and the peak is searched step by step to obtain the final ten results, and the results are statistically analyzed to obtain the histogram. Figure 3 (d) is when BATCH is 20, Figure 3 The data in (b) is filtered ten times and the peak is searched step by step to obtain the final ten results, and the results are statistically analyzed to obtain the histogram. It can be seen that IIR filtering can make the final result closer to the mean value of 6000. Compared with not filtering, directly counting the original data has a better effect.

[0045] Furthermore, the present invention also provides a data processing device for laser radar ranging, a filtering processing module, a first statistical module, a second statistical module, a third statistical module and a result output module, wherein:

[0046] The filtering processing module is used to filter the data with a bit width of W bits from the TDC using an IIR filter;

[0047] The first statistical module is used to count the first M bits of the W bits of data after filtering to obtain data T1 corresponding to the peak value, and a total of BATCH data are counted, where BATCH is a preset value;

[0048] The second statistical module is used to select the first M bits of data equal to T1 from the data output by TDC, count the middle M bits of these data, obtain the data T2 corresponding to the peak value, and count BATCH data in total;

[0049] The third statistical module is used to select the first 2M bits of data that are equal to [T1, T2] or [T1, T2]-1 from the data output by the TDC, count the last N bits of these data, obtain the data T3 corresponding to the peak value, and count 3*BATCH data in total;

[0050] The result output module is used to obtain the statistical result [T1, T2, T3];

[0051] Where W is the bit width of TDC, M and N are preset values, M and N must satisfy W=M+M+N, and BATCH is determined by the output frame rate of on-chip statistics and the frequency of TDC input data.

[0052] Furthermore, in the filtering processing module, filtering processing is used to eliminate noise from laser pulse width, SPAD jitter, and TDC circuit signal jitter.

[0053] Furthermore, in the filtering processing module, the time domain expression of the IIR filter used is: y[k]=0.875·y[k-1]+0.125·x[k].

[0054] Furthermore, the BATCH is determined by the following formula: data used for one statistic=BATCH+BATCH+3*BATCH=frequency of TDC input data / output frequency of the statistic module required by the system.

[0055] Furthermore, the value of W is 14, the value of M is 5, and the value of N is 4.

[0056] It will be easily understood by those skilled in the art that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A data processing method for laser radar ranging, characterized in that: include: (1) Using an IIR filter to filter the W-bit data from the TDC; (2) Count the first M bits of the W bits of data after filtering to obtain data T1 corresponding to the peak value, and count BATCH data in total, where BATCH is a preset value; (3) In the data output by TDC, select the data whose first M bits are equal to T1, count the middle M bits of these data, and obtain the data T2 corresponding to the peak value, and count BATCH data in total; (4) In the data output by TDC, select the data whose first 2M bits are equal to [T1, T2] or [T1, T2]-1, count the last N bits of these data, and obtain the data T3 corresponding to the peak value. A total of 3*BATCH data are counted; (5) The statistical result is [T1, T2, T3]; Where W is the bit width of TDC, M and N are preset values, M and N must satisfy W=M+M+N, and BATCH is determined by the output frame rate of on-chip statistics and the frequency of TDC input data.

2. The data processing method for laser radar ranging according to claim 1, characterized in that: In step (1), filtering is used to eliminate noise from laser pulse width, SPAD jitter, and TDC circuit signal jitter.

3. The data processing method for laser radar ranging according to claim 1 or 2, characterized in that: In step (1), the time domain expression of the IIR filter used is: y[k]=0.875·y[k-1]+0.125·x[k].

4. The data processing method for laser radar ranging according to claim 1 or 2, characterized in that: The BATCH is determined by the following formula: data used for one statistic=BATCH+BATCH+3*BATCH=frequency of TDC input data / output frequency of the statistic module required by the system.

5. The data processing method for laser radar ranging according to claim 1 or 2, characterized in that: The value of W is 14, the value of M is 5, and the value of N is 4.

6. A data processing device for laser radar ranging, characterized in that: A filtering processing module, a first statistical module, a second statistical module, a third statistical module and a result output module, wherein: The filtering processing module is used to filter the data with a bit width of W bits from the TDC using an IIR filter; The first statistical module is used to count the first M bits of the W bits of data after filtering to obtain data T1 corresponding to the peak value, and a total of BATCH data are counted, where BATCH is a preset value; The second statistical module is used to select the first M bits of data equal to T1 from the data output by TDC, count the middle M bits of these data, obtain the data T2 corresponding to the peak value, and count BATCH data in total; The third statistical module is used to select the first 2M bits of data that are equal to [T1, T2] or [T1, T2]-1 from the data output by the TDC, count the last N bits of these data, obtain the data T3 corresponding to the peak value, and count 3*BATCH data in total; The result output module is used to obtain the statistical result [T1, T2, T3]; Where W is the bit width of TDC, M and N are preset values, M and N must satisfy W=M+M+N, and BATCH is determined by the output frame rate of on-chip statistics and the frequency of TDC input data.

7. The data processing device for laser radar ranging according to claim 6, characterized in that: In the filtering processing module, filtering processing is used to eliminate noise from laser pulse width, SPAD jitter, and TDC circuit signal jitter.

8. The data processing device for laser radar ranging according to claim 6 or 7, characterized in that: In the filtering processing module, the time domain expression of the IIR filter used is: y[k]=0.875·y[k-1]+0.125·x[k].

9. The data processing device for laser radar ranging according to claim 6 or 7, characterized in that: The BATCH is determined by the following formula: data used for one statistic=BATCH+BATCH+3*BATCH=frequency of TDC input data / output frequency of the statistic module required by the system.

10. The data processing device for laser radar ranging according to claim 6 or 7, characterized in that: The value of W is 14, the value of M is 5, and the value of N is 4.

Images